messages
listlengths 3
3
|
|---|
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/025fa4bc16915ed6.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Butterfly</name>\n <bndbox>\n <xmin>0.15</xmin>\n <ymin>0.7</ymin>\n <xmax>0.31</xmax>\n <ymax>0.86</ymax>\n </bndbox>\n </object>\n <object>\n <name>Butterfly</name>\n <bndbox>\n <xmin>0.36</xmin>\n <ymin>0.26</ymin>\n <xmax>0.52</xmax>\n <ymax>0.51</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/05241c698a7771d1.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Boy</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.17</ymin>\n <xmax>0.51</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.59</xmin>\n <ymin>0.21</ymin>\n <xmax>0.63</xmax>\n <ymax>0.25</ymax>\n </bndbox>\n </object>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.62</xmin>\n <ymin>0.04</ymin>\n <xmax>0.68</xmax>\n <ymax>0.08</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.35</xmin>\n <ymin>0.22</ymin>\n <xmax>0.45</xmax>\n <ymax>0.43</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.52</xmin>\n <ymin>0.26</ymin>\n <xmax>0.61</xmax>\n <ymax>0.44</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/083fe185c33ff147.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Car</name>\n <bndbox>\n <xmin>0.05</xmin>\n <ymin>0.43</ymin>\n <xmax>0.21</xmax>\n <ymax>0.67</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/051210f6d3c8780a.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Fashion accessory</name>\n <bndbox>\n <xmin>0.86</xmin>\n <ymin>0.96</ymin>\n <xmax>0.89</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Fashion accessory</name>\n <bndbox>\n <xmin>0.93</xmin>\n <ymin>0.81</ymin>\n <xmax>1</xmax>\n <ymax>0.87</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/002509b57db1e7e7.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Man</name>\n <bndbox>\n <xmin>0.45</xmin>\n <ymin>0.05</ymin>\n <xmax>0.97</xmax>\n <ymax>0.98</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0089fa633ab4325d.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Pillow</name>\n <bndbox>\n <xmin>0.29</xmin>\n <ymin>0.6</ymin>\n <xmax>0.36</xmax>\n <ymax>0.7</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/00dba1c1bac3c761.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Woman</name>\n <bndbox>\n <xmin>0.19</xmin>\n <ymin>0.19</ymin>\n <xmax>0.84</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.23</ymin>\n <xmax>0.43</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.19</xmin>\n <ymin>0.59</ymin>\n <xmax>0.88</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/01df541bf57cf4fc.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Canary</name>\n <bndbox>\n <xmin>0.35</xmin>\n <ymin>0.26</ymin>\n <xmax>0.78</xmax>\n <ymax>0.78</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0776a456f8d65f9f.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Cowboy hat</name>\n <bndbox>\n <xmin>0.39</xmin>\n <ymin>0.08</ymin>\n <xmax>0.86</xmax>\n <ymax>0.2</ymax>\n </bndbox>\n </object>\n <object>\n <name>Sun hat</name>\n <bndbox>\n <xmin>0.38</xmin>\n <ymin>0.06</ymin>\n <xmax>0.85</xmax>\n <ymax>0.22</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/08186c4a4a775776.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Airplane</name>\n <bndbox>\n <xmin>0.09</xmin>\n <ymin>0.39</ymin>\n <xmax>0.51</xmax>\n <ymax>0.6</ymax>\n </bndbox>\n </object>\n <object>\n <name>Airplane</name>\n <bndbox>\n <xmin>0.46</xmin>\n <ymin>0.51</ymin>\n <xmax>0.94</xmax>\n <ymax>0.71</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/04cc2585f1f56e44.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Taco</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0</ymin>\n <xmax>0.99</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/05c52ccdb095f060.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Human hair</name>\n <bndbox>\n <xmin>0.5</xmin>\n <ymin>0.25</ymin>\n <xmax>0.64</xmax>\n <ymax>0.36</ymax>\n </bndbox>\n </object>\n <object>\n <name>Footwear</name>\n <bndbox>\n <xmin>0.76</xmin>\n <ymin>0.62</ymin>\n <xmax>0.9</xmax>\n <ymax>0.76</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/088edea3f8182502.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Football</name>\n <bndbox>\n <xmin>0.46</xmin>\n <ymin>0.84</ymin>\n <xmax>0.59</xmax>\n <ymax>0.96</ymax>\n </bndbox>\n </object>\n <object>\n <name>Fashion accessory</name>\n <bndbox>\n <xmin>0.34</xmin>\n <ymin>0.81</ymin>\n <xmax>0.46</xmax>\n <ymax>0.92</ymax>\n </bndbox>\n </object>\n <object>\n <name>Fashion accessory</name>\n <bndbox>\n <xmin>0.6</xmin>\n <ymin>0.85</ymin>\n <xmax>0.72</xmax>\n <ymax>0.92</ymax>\n </bndbox>\n </object>\n <object>\n <name>Fashion accessory</name>\n <bndbox>\n <xmin>0.63</xmin>\n <ymin>0.15</ymin>\n <xmax>0.73</xmax>\n <ymax>0.21</ymax>\n </bndbox>\n </object>\n <object>\n <name>Fashion accessory</name>\n <bndbox>\n <xmin>0.86</xmin>\n <ymin>0.87</ymin>\n <xmax>0.92</xmax>\n <ymax>0.96</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.38</xmin>\n <ymin>0.17</ymin>\n <xmax>0.45</xmax>\n <ymax>0.28</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.64</xmin>\n <ymin>0.17</ymin>\n <xmax>0.71</xmax>\n <ymax>0.28</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/03fcc4d4897ce9a0.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Vehicle registration plate</name>\n <bndbox>\n <xmin>0.96</xmin>\n <ymin>0.9</ymin>\n <xmax>1</xmax>\n <ymax>0.93</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/06985092821c60ba.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Skyscraper</name>\n <bndbox>\n <xmin>0.17</xmin>\n <ymin>0.08</ymin>\n <xmax>0.38</xmax>\n <ymax>0.42</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tower</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.39</ymin>\n <xmax>0.24</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.4</xmin>\n <ymin>0.89</ymin>\n <xmax>0.42</xmax>\n <ymax>0.95</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.46</xmin>\n <ymin>0.88</ymin>\n <xmax>0.48</xmax>\n <ymax>0.93</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/07aa572a1aef403e.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Mug</name>\n <bndbox>\n <xmin>0.23</xmin>\n <ymin>0.19</ymin>\n <xmax>0.78</xmax>\n <ymax>0.83</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/018fcbc2f5d0e11b.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Cat</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.11</ymin>\n <xmax>0.42</xmax>\n <ymax>0.79</ymax>\n </bndbox>\n </object>\n <object>\n <name>Cat</name>\n <bndbox>\n <xmin>0.47</xmin>\n <ymin>0.02</ymin>\n <xmax>0.51</xmax>\n <ymax>0.15</ymax>\n </bndbox>\n </object>\n <object>\n <name>Cat</name>\n <bndbox>\n <xmin>0.56</xmin>\n <ymin>0.04</ymin>\n <xmax>1</xmax>\n <ymax>0.84</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/012aa90ff6e08433.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Sun hat</name>\n <bndbox>\n <xmin>0.15</xmin>\n <ymin>0.1</ymin>\n <xmax>0.59</xmax>\n <ymax>0.54</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0128de6e72f816ef.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Tree</name>\n <bndbox>\n <xmin>0.02</xmin>\n <ymin>0.03</ymin>\n <xmax>0.35</xmax>\n <ymax>0.78</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0791e242096f7c36.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Toy</name>\n <bndbox>\n <xmin>0.01</xmin>\n <ymin>0.3</ymin>\n <xmax>0.09</xmax>\n <ymax>0.39</ymax>\n </bndbox>\n </object>\n <object>\n <name>Toy</name>\n <bndbox>\n <xmin>0.33</xmin>\n <ymin>0.68</ymin>\n <xmax>0.38</xmax>\n <ymax>0.78</ymax>\n </bndbox>\n </object>\n <object>\n <name>Toy</name>\n <bndbox>\n <xmin>0.47</xmin>\n <ymin>0.44</ymin>\n <xmax>0.8</xmax>\n <ymax>0.79</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/00abb2ce20182df8.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.48</xmin>\n <ymin>0.31</ymin>\n <xmax>0.56</xmax>\n <ymax>0.48</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.67</xmin>\n <ymin>0.33</ymin>\n <xmax>0.77</xmax>\n <ymax>0.51</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/005d044af0077dd5.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Guitar</name>\n <bndbox>\n <xmin>0.89</xmin>\n <ymin>0.5</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/013ecdbd4eef7d55.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Tree</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.07</ymin>\n <xmax>0.1</xmax>\n <ymax>0.36</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tree</name>\n <bndbox>\n <xmin>0.14</xmin>\n <ymin>0</ymin>\n <xmax>0.3</xmax>\n <ymax>0.61</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/04bf850cb891592a.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.49</xmin>\n <ymin>0.62</ymin>\n <xmax>0.71</xmax>\n <ymax>0.91</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.01</xmin>\n <ymin>0.81</ymin>\n <xmax>0.04</xmax>\n <ymax>0.89</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.01</xmin>\n <ymin>0</ymin>\n <xmax>0.04</xmax>\n <ymax>0.06</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.07</xmin>\n <ymin>0.4</ymin>\n <xmax>0.11</xmax>\n <ymax>0.47</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.14</xmin>\n <ymin>0.23</ymin>\n <xmax>0.18</xmax>\n <ymax>0.29</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.14</xmin>\n <ymin>0.68</ymin>\n <xmax>0.2</xmax>\n <ymax>0.77</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.19</xmin>\n <ymin>0.35</ymin>\n <xmax>0.24</xmax>\n <ymax>0.41</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.24</xmin>\n <ymin>0.44</ymin>\n <xmax>0.3</xmax>\n <ymax>0.51</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.25</xmin>\n <ymin>0.03</ymin>\n <xmax>0.29</xmax>\n <ymax>0.09</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.26</xmin>\n <ymin>0.21</ymin>\n <xmax>0.31</xmax>\n <ymax>0.29</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.34</xmin>\n <ymin>0.09</ymin>\n <xmax>0.39</xmax>\n <ymax>0.17</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.37</xmin>\n <ymin>0.3</ymin>\n <xmax>0.42</xmax>\n <ymax>0.36</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.39</xmin>\n <ymin>0.45</ymin>\n <xmax>0.43</xmax>\n <ymax>0.53</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.41</xmin>\n <ymin>0.17</ymin>\n <xmax>0.46</xmax>\n <ymax>0.23</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.46</xmin>\n <ymin>0.03</ymin>\n <xmax>0.51</xmax>\n <ymax>0.09</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.5</xmin>\n <ymin>0.32</ymin>\n <xmax>0.55</xmax>\n <ymax>0.39</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.56</xmin>\n <ymin>0.18</ymin>\n <xmax>0.61</xmax>\n <ymax>0.24</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.61</xmin>\n <ymin>0.01</ymin>\n <xmax>0.66</xmax>\n <ymax>0.08</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.61</xmin>\n <ymin>0.48</ymin>\n <xmax>0.67</xmax>\n <ymax>0.56</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.67</xmin>\n <ymin>0.23</ymin>\n <xmax>0.72</xmax>\n <ymax>0.3</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.8</xmin>\n <ymin>0.42</ymin>\n <xmax>0.84</xmax>\n <ymax>0.5</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.8</xmin>\n <ymin>0</ymin>\n <xmax>0.85</xmax>\n <ymax>0.06</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.89</xmin>\n <ymin>0.33</ymin>\n <xmax>0.94</xmax>\n <ymax>0.39</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/013ca61b36619a59.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Plant</name>\n <bndbox>\n <xmin>0.81</xmin>\n <ymin>0.73</ymin>\n <xmax>0.95</xmax>\n <ymax>0.94</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.56</xmin>\n <ymin>0.29</ymin>\n <xmax>0.65</xmax>\n <ymax>0.42</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/007575a3d2399034.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.49</xmin>\n <ymin>0.14</ymin>\n <xmax>0.75</xmax>\n <ymax>0.34</ymax>\n </bndbox>\n </object>\n <object>\n <name>Boy</name>\n <bndbox>\n <xmin>0.03</xmin>\n <ymin>0.46</ymin>\n <xmax>0.54</xmax>\n <ymax>0.93</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.15</xmin>\n <ymin>0.45</ymin>\n <xmax>0.29</xmax>\n <ymax>0.56</ymax>\n </bndbox>\n </object>\n <object>\n <name>Sports uniform</name>\n <bndbox>\n <xmin>0.05</xmin>\n <ymin>0.53</ymin>\n <xmax>0.53</xmax>\n <ymax>0.82</ymax>\n </bndbox>\n </object>\n <object>\n <name>Sports uniform</name>\n <bndbox>\n <xmin>0.21</xmin>\n <ymin>0.35</ymin>\n <xmax>0.92</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/028ac57e651072fc.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Sports uniform</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.18</ymin>\n <xmax>0.65</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/001bcbe345f8b23b.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Sports uniform</name>\n <bndbox>\n <xmin>0.43</xmin>\n <ymin>0.21</ymin>\n <xmax>0.82</xmax>\n <ymax>0.79</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/02d7141f40325a6b.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Human eye</name>\n <bndbox>\n <xmin>0.29</xmin>\n <ymin>0.33</ymin>\n <xmax>0.41</xmax>\n <ymax>0.37</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human eye</name>\n <bndbox>\n <xmin>0.5</xmin>\n <ymin>0.3</ymin>\n <xmax>0.64</xmax>\n <ymax>0.36</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/085afc7b51d2362c.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Boy</name>\n <bndbox>\n <xmin>0.07</xmin>\n <ymin>0.11</ymin>\n <xmax>1</xmax>\n <ymax>0.96</ymax>\n </bndbox>\n </object>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.11</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0191fc13b7614846.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Cat</name>\n <bndbox>\n <xmin>0.59</xmin>\n <ymin>0.11</ymin>\n <xmax>0.92</xmax>\n <ymax>0.32</ymax>\n </bndbox>\n </object>\n <object>\n <name>Chicken</name>\n <bndbox>\n <xmin>0.11</xmin>\n <ymin>0.54</ymin>\n <xmax>0.34</xmax>\n <ymax>0.96</ymax>\n </bndbox>\n </object>\n <object>\n <name>Chicken</name>\n <bndbox>\n <xmin>0.43</xmin>\n <ymin>0.47</ymin>\n <xmax>0.55</xmax>\n <ymax>0.87</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/049149cbd7ecbd7d.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Radish</name>\n <bndbox>\n <xmin>0.63</xmin>\n <ymin>0</ymin>\n <xmax>0.72</xmax>\n <ymax>0.11</ymax>\n </bndbox>\n </object>\n <object>\n <name>Radish</name>\n <bndbox>\n <xmin>0.7</xmin>\n <ymin>0.16</ymin>\n <xmax>0.81</xmax>\n <ymax>0.31</ymax>\n </bndbox>\n </object>\n <object>\n <name>Radish</name>\n <bndbox>\n <xmin>0.76</xmin>\n <ymin>0.4</ymin>\n <xmax>0.88</xmax>\n <ymax>0.58</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tart</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.01</ymin>\n <xmax>1</xmax>\n <ymax>0.98</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tart</name>\n <bndbox>\n <xmin>0.3</xmin>\n <ymin>0.47</ymin>\n <xmax>0.68</xmax>\n <ymax>0.97</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0033d73d085b10f6.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Trousers</name>\n <bndbox>\n <xmin>0.52</xmin>\n <ymin>0.52</ymin>\n <xmax>0.63</xmax>\n <ymax>0.63</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/01b184b9ec0d54de.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Airplane</name>\n <bndbox>\n <xmin>0.42</xmin>\n <ymin>0</ymin>\n <xmax>1</xmax>\n <ymax>0.82</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/001c40589d0bfd61.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Toy</name>\n <bndbox>\n <xmin>0.03</xmin>\n <ymin>0.01</ymin>\n <xmax>0.97</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Land vehicle</name>\n <bndbox>\n <xmin>0.03</xmin>\n <ymin>0</ymin>\n <xmax>0.98</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Wheel</name>\n <bndbox>\n <xmin>0.31</xmin>\n <ymin>0.8</ymin>\n <xmax>0.55</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tire</name>\n <bndbox>\n <xmin>0.31</xmin>\n <ymin>0.8</ymin>\n <xmax>0.54</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/00b9c2e27130cb26.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Bird</name>\n <bndbox>\n <xmin>0.35</xmin>\n <ymin>0.3</ymin>\n <xmax>0.6</xmax>\n <ymax>0.95</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/04ae96301a6b8c71.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.05</xmin>\n <ymin>0.51</ymin>\n <xmax>0.15</xmax>\n <ymax>0.73</ymax>\n </bndbox>\n </object>\n <object>\n <name>Woman</name>\n <bndbox>\n <xmin>0.27</xmin>\n <ymin>0.15</ymin>\n <xmax>0.4</xmax>\n <ymax>0.35</ymax>\n </bndbox>\n </object>\n <object>\n <name>Man</name>\n <bndbox>\n <xmin>0.02</xmin>\n <ymin>0.85</ymin>\n <xmax>0.23</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.27</xmin>\n <ymin>0.22</ymin>\n <xmax>0.37</xmax>\n <ymax>0.34</ymax>\n </bndbox>\n </object>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.39</xmin>\n <ymin>0.4</ymin>\n <xmax>0.62</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.51</xmin>\n <ymin>0.64</ymin>\n <xmax>0.58</xmax>\n <ymax>0.79</ymax>\n </bndbox>\n </object>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.67</xmin>\n <ymin>0.77</ymin>\n <xmax>0.83</xmax>\n <ymax>0.99</ymax>\n </bndbox>\n </object>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.8</xmin>\n <ymin>0.63</ymin>\n <xmax>0.85</xmax>\n <ymax>0.73</ymax>\n </bndbox>\n </object>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.92</xmin>\n <ymin>0.77</ymin>\n <xmax>1</xmax>\n <ymax>0.93</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.09</xmin>\n <ymin>0.51</ymin>\n <xmax>0.11</xmax>\n <ymax>0.53</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.31</xmin>\n <ymin>0.17</ymin>\n <xmax>0.35</xmax>\n <ymax>0.22</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.32</xmin>\n <ymin>0.73</ymin>\n <xmax>0.35</xmax>\n <ymax>0.77</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.56</xmin>\n <ymin>0.75</ymin>\n <xmax>0.6</xmax>\n <ymax>0.81</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.64</xmin>\n <ymin>0.75</ymin>\n <xmax>0.66</xmax>\n <ymax>0.78</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.66</xmin>\n <ymin>0.69</ymin>\n <xmax>0.68</xmax>\n <ymax>0.74</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.74</xmin>\n <ymin>0.74</ymin>\n <xmax>0.78</xmax>\n <ymax>0.81</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.8</xmin>\n <ymin>0.74</ymin>\n <xmax>0.83</xmax>\n <ymax>0.77</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.82</xmin>\n <ymin>0.62</ymin>\n <xmax>0.84</xmax>\n <ymax>0.64</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.86</xmin>\n <ymin>0.75</ymin>\n <xmax>0.89</xmax>\n <ymax>0.79</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.93</xmin>\n <ymin>0.7</ymin>\n <xmax>0.94</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.95</xmin>\n <ymin>0.75</ymin>\n <xmax>0.98</xmax>\n <ymax>0.79</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.97</xmin>\n <ymin>0.67</ymin>\n <xmax>0.98</xmax>\n <ymax>0.69</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/020c39b5c286b515.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Man</name>\n <bndbox>\n <xmin>0.76</xmin>\n <ymin>0.17</ymin>\n <xmax>0.97</xmax>\n <ymax>0.96</ymax>\n </bndbox>\n </object>\n <object>\n <name>Footwear</name>\n <bndbox>\n <xmin>0.88</xmin>\n <ymin>0.84</ymin>\n <xmax>0.96</xmax>\n <ymax>0.94</ymax>\n </bndbox>\n </object>\n <object>\n <name>Man</name>\n <bndbox>\n <xmin>0.06</xmin>\n <ymin>0.27</ymin>\n <xmax>0.35</xmax>\n <ymax>0.95</ymax>\n </bndbox>\n </object>\n <object>\n <name>Man</name>\n <bndbox>\n <xmin>0.29</xmin>\n <ymin>0.25</ymin>\n <xmax>0.56</xmax>\n <ymax>0.96</ymax>\n </bndbox>\n </object>\n <object>\n <name>Sports equipment</name>\n <bndbox>\n <xmin>0.2</xmin>\n <ymin>0.92</ymin>\n <xmax>0.26</xmax>\n <ymax>0.95</ymax>\n </bndbox>\n </object>\n <object>\n <name>Footwear</name>\n <bndbox>\n <xmin>0.46</xmin>\n <ymin>0.85</ymin>\n <xmax>0.52</xmax>\n <ymax>0.95</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/00379a50c35b6527.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Snack</name>\n <bndbox>\n <xmin>0.22</xmin>\n <ymin>0</ymin>\n <xmax>0.71</xmax>\n <ymax>0.99</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/00f640154db6e39b.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Bicycle</name>\n <bndbox>\n <xmin>0.16</xmin>\n <ymin>0.57</ymin>\n <xmax>0.85</xmax>\n <ymax>0.83</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/06df106a13858bd8.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.13</xmin>\n <ymin>0</ymin>\n <xmax>0.19</xmax>\n <ymax>0.11</ymax>\n </bndbox>\n </object>\n <object>\n <name>Motorcycle</name>\n <bndbox>\n <xmin>0.02</xmin>\n <ymin>0</ymin>\n <xmax>0.97</xmax>\n <ymax>0.96</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/08cc414fae3558d6.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Tree</name>\n <bndbox>\n <xmin>0.07</xmin>\n <ymin>0.75</ymin>\n <xmax>0.26</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tree</name>\n <bndbox>\n <xmin>0.35</xmin>\n <ymin>0.5</ymin>\n <xmax>0.45</xmax>\n <ymax>0.69</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tree</name>\n <bndbox>\n <xmin>0.61</xmin>\n <ymin>0.55</ymin>\n <xmax>0.71</xmax>\n <ymax>0.94</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tree</name>\n <bndbox>\n <xmin>0.82</xmin>\n <ymin>0.53</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/039164da2237f6e3.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Tree</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/05b149677d7ec9aa.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Footwear</name>\n <bndbox>\n <xmin>0.4</xmin>\n <ymin>0.45</ymin>\n <xmax>0.45</xmax>\n <ymax>0.49</ymax>\n </bndbox>\n </object>\n <object>\n <name>Footwear</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.46</ymin>\n <xmax>0.58</xmax>\n <ymax>0.49</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0459aac9e7d7863f.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Car</name>\n <bndbox>\n <xmin>0.01</xmin>\n <ymin>0.28</ymin>\n <xmax>0.94</xmax>\n <ymax>0.79</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/015cc4d3633407c7.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Drink</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.03</ymin>\n <xmax>1</xmax>\n <ymax>0.59</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0771904ed1d8cb0c.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Footwear</name>\n <bndbox>\n <xmin>0.28</xmin>\n <ymin>0.66</ymin>\n <xmax>0.35</xmax>\n <ymax>0.77</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/01459f9d3724adf8.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Taxi</name>\n <bndbox>\n <xmin>0.25</xmin>\n <ymin>0.59</ymin>\n <xmax>0.43</xmax>\n <ymax>0.66</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/02d857b453d230b3.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Bicycle helmet</name>\n <bndbox>\n <xmin>0.07</xmin>\n <ymin>0.42</ymin>\n <xmax>0.11</xmax>\n <ymax>0.53</ymax>\n </bndbox>\n </object>\n <object>\n <name>Bicycle helmet</name>\n <bndbox>\n <xmin>0.21</xmin>\n <ymin>0.39</ymin>\n <xmax>0.24</xmax>\n <ymax>0.46</ymax>\n </bndbox>\n </object>\n <object>\n <name>Bicycle helmet</name>\n <bndbox>\n <xmin>0.28</xmin>\n <ymin>0.43</ymin>\n <xmax>0.32</xmax>\n <ymax>0.48</ymax>\n </bndbox>\n </object>\n <object>\n <name>Bicycle helmet</name>\n <bndbox>\n <xmin>0.35</xmin>\n <ymin>0.42</ymin>\n <xmax>0.39</xmax>\n <ymax>0.48</ymax>\n </bndbox>\n </object>\n <object>\n <name>Bicycle helmet</name>\n <bndbox>\n <xmin>0.42</xmin>\n <ymin>0.42</ymin>\n <xmax>0.45</xmax>\n <ymax>0.47</ymax>\n </bndbox>\n </object>\n <object>\n <name>Bicycle helmet</name>\n <bndbox>\n <xmin>0.45</xmin>\n <ymin>0.42</ymin>\n <xmax>0.47</xmax>\n <ymax>0.47</ymax>\n </bndbox>\n </object>\n <object>\n <name>Bicycle helmet</name>\n <bndbox>\n <xmin>0.48</xmin>\n <ymin>0.44</ymin>\n <xmax>0.5</xmax>\n <ymax>0.49</ymax>\n </bndbox>\n </object>\n <object>\n <name>Bicycle helmet</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.43</ymin>\n <xmax>0.57</xmax>\n <ymax>0.48</ymax>\n </bndbox>\n </object>\n <object>\n <name>Bicycle helmet</name>\n <bndbox>\n <xmin>0.56</xmin>\n <ymin>0.4</ymin>\n <xmax>0.59</xmax>\n <ymax>0.47</ymax>\n </bndbox>\n </object>\n <object>\n <name>Bicycle helmet</name>\n <bndbox>\n <xmin>0.61</xmin>\n <ymin>0.42</ymin>\n <xmax>0.63</xmax>\n <ymax>0.5</ymax>\n </bndbox>\n </object>\n <object>\n <name>Bicycle helmet</name>\n <bndbox>\n <xmin>0.64</xmin>\n <ymin>0.42</ymin>\n <xmax>0.66</xmax>\n <ymax>0.48</ymax>\n </bndbox>\n </object>\n <object>\n <name>Bicycle helmet</name>\n <bndbox>\n <xmin>0.67</xmin>\n <ymin>0.4</ymin>\n <xmax>0.69</xmax>\n <ymax>0.44</ymax>\n </bndbox>\n </object>\n <object>\n <name>Bicycle helmet</name>\n <bndbox>\n <xmin>0.7</xmin>\n <ymin>0.37</ymin>\n <xmax>0.72</xmax>\n <ymax>0.41</ymax>\n </bndbox>\n </object>\n <object>\n <name>Bicycle helmet</name>\n <bndbox>\n <xmin>0.71</xmin>\n <ymin>0.42</ymin>\n <xmax>0.73</xmax>\n <ymax>0.47</ymax>\n </bndbox>\n </object>\n <object>\n <name>Bicycle helmet</name>\n <bndbox>\n <xmin>0.77</xmin>\n <ymin>0.44</ymin>\n <xmax>0.78</xmax>\n <ymax>0.49</ymax>\n </bndbox>\n </object>\n <object>\n <name>Bicycle helmet</name>\n <bndbox>\n <xmin>0.78</xmin>\n <ymin>0.41</ymin>\n <xmax>0.8</xmax>\n <ymax>0.45</ymax>\n </bndbox>\n </object>\n <object>\n <name>Bicycle helmet</name>\n <bndbox>\n <xmin>0.83</xmin>\n <ymin>0.41</ymin>\n <xmax>0.85</xmax>\n <ymax>0.45</ymax>\n </bndbox>\n </object>\n <object>\n <name>Bicycle helmet</name>\n <bndbox>\n <xmin>0.91</xmin>\n <ymin>0.42</ymin>\n <xmax>0.93</xmax>\n <ymax>0.47</ymax>\n </bndbox>\n </object>\n <object>\n <name>Bicycle helmet</name>\n <bndbox>\n <xmin>0.96</xmin>\n <ymin>0.41</ymin>\n <xmax>0.98</xmax>\n <ymax>0.44</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/02ad52d2f57ff4e9.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.31</xmin>\n <ymin>0.28</ymin>\n <xmax>0.42</xmax>\n <ymax>0.78</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/06994fb3e2879945.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.38</xmin>\n <ymin>0.18</ymin>\n <xmax>0.87</xmax>\n <ymax>0.99</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/006b6bdec2e7240b.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Bus</name>\n <bndbox>\n <xmin>0.13</xmin>\n <ymin>0.29</ymin>\n <xmax>0.93</xmax>\n <ymax>0.87</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0854552d31c4a719.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Sculpture</name>\n <bndbox>\n <xmin>0.17</xmin>\n <ymin>0.23</ymin>\n <xmax>0.83</xmax>\n <ymax>0.79</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/07d5e5656c0965c9.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Land vehicle</name>\n <bndbox>\n <xmin>0.84</xmin>\n <ymin>0.12</ymin>\n <xmax>0.99</xmax>\n <ymax>0.21</ymax>\n </bndbox>\n </object>\n <object>\n <name>Waste container</name>\n <bndbox>\n <xmin>0.51</xmin>\n <ymin>0.18</ymin>\n <xmax>0.62</xmax>\n <ymax>0.33</ymax>\n </bndbox>\n </object>\n <object>\n <name>Land vehicle</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.13</ymin>\n <xmax>0.59</xmax>\n <ymax>0.15</ymax>\n </bndbox>\n </object>\n <object>\n <name>Waste container</name>\n <bndbox>\n <xmin>0.02</xmin>\n <ymin>0.4</ymin>\n <xmax>0.55</xmax>\n <ymax>0.76</ymax>\n </bndbox>\n </object>\n <object>\n <name>Waste container</name>\n <bndbox>\n <xmin>0.39</xmin>\n <ymin>0.22</ymin>\n <xmax>0.54</xmax>\n <ymax>0.3</ymax>\n </bndbox>\n </object>\n <object>\n <name>Waste container</name>\n <bndbox>\n <xmin>0.49</xmin>\n <ymin>0.14</ymin>\n <xmax>0.52</xmax>\n <ymax>0.18</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/08c0101dc22906ab.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Man</name>\n <bndbox>\n <xmin>0.32</xmin>\n <ymin>0.32</ymin>\n <xmax>0.79</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Jeans</name>\n <bndbox>\n <xmin>0.4</xmin>\n <ymin>0.84</ymin>\n <xmax>0.74</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.45</xmin>\n <ymin>0.34</ymin>\n <xmax>0.58</xmax>\n <ymax>0.46</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0200f89db292a722.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.07</xmin>\n <ymin>0.41</ymin>\n <xmax>0.4</xmax>\n <ymax>0.96</ymax>\n </bndbox>\n </object>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.37</xmin>\n <ymin>0.35</ymin>\n <xmax>0.69</xmax>\n <ymax>0.91</ymax>\n </bndbox>\n </object>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.68</xmin>\n <ymin>0.34</ymin>\n <xmax>0.89</xmax>\n <ymax>0.83</ymax>\n </bndbox>\n </object>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.77</xmin>\n <ymin>0.79</ymin>\n <xmax>0.89</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.24</xmin>\n <ymin>0.22</ymin>\n <xmax>0.35</xmax>\n <ymax>0.41</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.47</xmin>\n <ymin>0.2</ymin>\n <xmax>0.58</xmax>\n <ymax>0.39</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.59</xmin>\n <ymin>0.11</ymin>\n <xmax>0.62</xmax>\n <ymax>0.15</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.68</xmin>\n <ymin>0.19</ymin>\n <xmax>0.77</xmax>\n <ymax>0.35</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.77</xmin>\n <ymin>0.06</ymin>\n <xmax>0.8</xmax>\n <ymax>0.12</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/083d6faa46f87483.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Airplane</name>\n <bndbox>\n <xmin>0.04</xmin>\n <ymin>0.16</ymin>\n <xmax>0.95</xmax>\n <ymax>0.83</ymax>\n </bndbox>\n </object>\n <object>\n <name>Airplane</name>\n <bndbox>\n <xmin>0.94</xmin>\n <ymin>0.52</ymin>\n <xmax>1</xmax>\n <ymax>0.62</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/08751ea3166750c6.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Tree</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0</ymin>\n <xmax>0.44</xmax>\n <ymax>0.95</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tree</name>\n <bndbox>\n <xmin>0.35</xmin>\n <ymin>0.02</ymin>\n <xmax>0.78</xmax>\n <ymax>0.78</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/002370f8520726aa.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Food</name>\n <bndbox>\n <xmin>0.04</xmin>\n <ymin>0.09</ymin>\n <xmax>0.94</xmax>\n <ymax>0.92</ymax>\n </bndbox>\n </object>\n <object>\n <name>Food</name>\n <bndbox>\n <xmin>0.21</xmin>\n <ymin>0.65</ymin>\n <xmax>0.63</xmax>\n <ymax>0.92</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/066565caddb3f0c0.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Human body</name>\n <bndbox>\n <xmin>0.81</xmin>\n <ymin>0.02</ymin>\n <xmax>0.99</xmax>\n <ymax>0.43</ymax>\n </bndbox>\n </object>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.75</xmin>\n <ymin>0.47</ymin>\n <xmax>0.97</xmax>\n <ymax>0.99</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human hand</name>\n <bndbox>\n <xmin>0.36</xmin>\n <ymin>0.19</ymin>\n <xmax>0.47</xmax>\n <ymax>0.34</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human hand</name>\n <bndbox>\n <xmin>0.85</xmin>\n <ymin>0.07</ymin>\n <xmax>0.93</xmax>\n <ymax>0.15</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/02780556015e6c80.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Girl</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0</ymin>\n <xmax>0.8</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/02d1f969683a52e9.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.4</ymin>\n <xmax>0.64</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.46</xmin>\n <ymin>0.07</ymin>\n <xmax>0.62</xmax>\n <ymax>0.29</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.53</xmin>\n <ymin>0.21</ymin>\n <xmax>0.81</xmax>\n <ymax>0.58</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/067c57199b6858c1.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.37</xmin>\n <ymin>0.77</ymin>\n <xmax>0.42</xmax>\n <ymax>0.96</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.68</ymin>\n <xmax>0.02</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.02</xmin>\n <ymin>0.24</ymin>\n <xmax>0.04</xmax>\n <ymax>0.29</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.02</xmin>\n <ymin>0.67</ymin>\n <xmax>0.06</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.03</xmin>\n <ymin>0.69</ymin>\n <xmax>0.09</xmax>\n <ymax>0.93</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.05</xmin>\n <ymin>0.44</ymin>\n <xmax>0.11</xmax>\n <ymax>0.56</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.08</xmin>\n <ymin>0.7</ymin>\n <xmax>0.13</xmax>\n <ymax>0.92</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.69</ymin>\n <xmax>0.17</xmax>\n <ymax>0.75</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.13</xmin>\n <ymin>0.65</ymin>\n <xmax>0.16</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.13</xmin>\n <ymin>0.75</ymin>\n <xmax>0.17</xmax>\n <ymax>0.94</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.15</xmin>\n <ymin>0.26</ymin>\n <xmax>0.17</xmax>\n <ymax>0.34</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.18</xmin>\n <ymin>0.73</ymin>\n <xmax>0.21</xmax>\n <ymax>0.92</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.22</xmin>\n <ymin>0.41</ymin>\n <xmax>0.29</xmax>\n <ymax>0.52</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.23</xmin>\n <ymin>0.73</ymin>\n <xmax>0.26</xmax>\n <ymax>0.92</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.26</xmin>\n <ymin>0.73</ymin>\n <xmax>0.3</xmax>\n <ymax>0.94</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.31</xmin>\n <ymin>0.29</ymin>\n <xmax>0.33</xmax>\n <ymax>0.35</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.31</xmin>\n <ymin>0.72</ymin>\n <xmax>0.35</xmax>\n <ymax>0.96</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.34</xmin>\n <ymin>0.28</ymin>\n <xmax>0.35</xmax>\n <ymax>0.35</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.35</xmin>\n <ymin>0.39</ymin>\n <xmax>0.38</xmax>\n <ymax>0.47</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.37</xmin>\n <ymin>0.28</ymin>\n <xmax>0.38</xmax>\n <ymax>0.34</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.38</xmin>\n <ymin>0.36</ymin>\n <xmax>0.4</xmax>\n <ymax>0.43</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.4</xmin>\n <ymin>0.3</ymin>\n <xmax>0.42</xmax>\n <ymax>0.38</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.47</xmin>\n <ymin>0.29</ymin>\n <xmax>0.49</xmax>\n <ymax>0.37</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.48</xmin>\n <ymin>0.75</ymin>\n <xmax>0.54</xmax>\n <ymax>0.96</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.5</xmin>\n <ymin>0.29</ymin>\n <xmax>0.54</xmax>\n <ymax>0.38</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.51</xmin>\n <ymin>0.39</ymin>\n <xmax>0.54</xmax>\n <ymax>0.47</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.54</xmin>\n <ymin>0.76</ymin>\n <xmax>0.6</xmax>\n <ymax>0.97</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.54</xmin>\n <ymin>0.36</ymin>\n <xmax>0.56</xmax>\n <ymax>0.43</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.57</xmin>\n <ymin>0.31</ymin>\n <xmax>0.6</xmax>\n <ymax>0.38</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.6</xmin>\n <ymin>0.77</ymin>\n <xmax>0.66</xmax>\n <ymax>0.99</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.61</xmin>\n <ymin>0.35</ymin>\n <xmax>0.65</xmax>\n <ymax>0.42</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.61</xmin>\n <ymin>0.58</ymin>\n <xmax>0.67</xmax>\n <ymax>0.69</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.67</xmin>\n <ymin>0.36</ymin>\n <xmax>0.7</xmax>\n <ymax>0.41</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.68</xmin>\n <ymin>0.79</ymin>\n <xmax>0.74</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.72</xmin>\n <ymin>0.32</ymin>\n <xmax>0.75</xmax>\n <ymax>0.41</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.73</xmin>\n <ymin>0.33</ymin>\n <xmax>0.77</xmax>\n <ymax>0.4</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.75</xmin>\n <ymin>0.81</ymin>\n <xmax>0.8</xmax>\n <ymax>0.99</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.75</xmin>\n <ymin>0.32</ymin>\n <xmax>0.78</xmax>\n <ymax>0.4</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.76</xmin>\n <ymin>0.5</ymin>\n <xmax>0.8</xmax>\n <ymax>0.61</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.78</xmin>\n <ymin>0.33</ymin>\n <xmax>0.8</xmax>\n <ymax>0.4</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.79</xmin>\n <ymin>0.34</ymin>\n <xmax>0.81</xmax>\n <ymax>0.41</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.8</xmin>\n <ymin>0.82</ymin>\n <xmax>0.85</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.8</xmin>\n <ymin>0.33</ymin>\n <xmax>0.83</xmax>\n <ymax>0.42</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.84</xmin>\n <ymin>0.33</ymin>\n <xmax>0.86</xmax>\n <ymax>0.42</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.86</xmin>\n <ymin>0.46</ymin>\n <xmax>0.89</xmax>\n <ymax>0.57</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.88</xmin>\n <ymin>0.33</ymin>\n <xmax>0.9</xmax>\n <ymax>0.42</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.88</xmin>\n <ymin>0.81</ymin>\n <xmax>0.94</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.9</xmin>\n <ymin>0.34</ymin>\n <xmax>0.92</xmax>\n <ymax>0.41</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.92</xmin>\n <ymin>0.35</ymin>\n <xmax>0.94</xmax>\n <ymax>0.43</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.94</xmin>\n <ymin>0.8</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.96</xmin>\n <ymin>0.44</ymin>\n <xmax>1</xmax>\n <ymax>0.53</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.97</xmin>\n <ymin>0.84</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Football helmet</name>\n <bndbox>\n <xmin>0.32</xmin>\n <ymin>0.73</ymin>\n <xmax>0.34</xmax>\n <ymax>0.75</ymax>\n </bndbox>\n </object>\n <object>\n <name>Football helmet</name>\n <bndbox>\n <xmin>0.37</xmin>\n <ymin>0.38</ymin>\n <xmax>0.38</xmax>\n <ymax>0.4</ymax>\n </bndbox>\n </object>\n <object>\n <name>Football helmet</name>\n <bndbox>\n <xmin>0.53</xmin>\n <ymin>0.38</ymin>\n <xmax>0.54</xmax>\n <ymax>0.4</ymax>\n </bndbox>\n </object>\n <object>\n <name>Football helmet</name>\n <bndbox>\n <xmin>0.97</xmin>\n <ymin>0.44</ymin>\n <xmax>0.99</xmax>\n <ymax>0.47</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/02a1b50f79e55984.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Street light</name>\n <bndbox>\n <xmin>0.71</xmin>\n <ymin>0.01</ymin>\n <xmax>0.75</xmax>\n <ymax>0.48</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/01dc72d4f485375e.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Cucumber</name>\n <bndbox>\n <xmin>0.01</xmin>\n <ymin>0.11</ymin>\n <xmax>1</xmax>\n <ymax>0.98</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/04071bd38a9ee6dc.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Skyscraper</name>\n <bndbox>\n <xmin>0.26</xmin>\n <ymin>0.08</ymin>\n <xmax>0.73</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/086b44c50dbc2f4c.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.06</xmin>\n <ymin>0.49</ymin>\n <xmax>0.38</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.13</xmin>\n <ymin>0.49</ymin>\n <xmax>0.28</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.25</xmin>\n <ymin>0.47</ymin>\n <xmax>0.34</xmax>\n <ymax>0.68</ymax>\n </bndbox>\n </object>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.35</xmin>\n <ymin>0.42</ymin>\n <xmax>0.62</xmax>\n <ymax>0.75</ymax>\n </bndbox>\n </object>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.57</xmin>\n <ymin>0.47</ymin>\n <xmax>0.66</xmax>\n <ymax>0.89</ymax>\n </bndbox>\n </object>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.64</xmin>\n <ymin>0.44</ymin>\n <xmax>0.81</xmax>\n <ymax>0.85</ymax>\n </bndbox>\n </object>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.65</xmin>\n <ymin>0.83</ymin>\n <xmax>0.78</xmax>\n <ymax>0.98</ymax>\n </bndbox>\n </object>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.74</xmin>\n <ymin>0.47</ymin>\n <xmax>1</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.79</xmin>\n <ymin>0.83</ymin>\n <xmax>0.99</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.69</xmin>\n <ymin>0.34</ymin>\n <xmax>0.76</xmax>\n <ymax>0.45</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/08209bc45e07c4dd.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>House</name>\n <bndbox>\n <xmin>0.67</xmin>\n <ymin>0.45</ymin>\n <xmax>0.94</xmax>\n <ymax>0.64</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/017b5c5c72bed7cc.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Fast food</name>\n <bndbox>\n <xmin>0.06</xmin>\n <ymin>0.61</ymin>\n <xmax>0.21</xmax>\n <ymax>0.68</ymax>\n </bndbox>\n </object>\n <object>\n <name>Fast food</name>\n <bndbox>\n <xmin>0.08</xmin>\n <ymin>0.95</ymin>\n <xmax>0.38</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Fast food</name>\n <bndbox>\n <xmin>0.11</xmin>\n <ymin>0.76</ymin>\n <xmax>0.4</xmax>\n <ymax>0.9</ymax>\n </bndbox>\n </object>\n <object>\n <name>Fast food</name>\n <bndbox>\n <xmin>0.11</xmin>\n <ymin>0.69</ymin>\n <xmax>0.32</xmax>\n <ymax>0.77</ymax>\n </bndbox>\n </object>\n <object>\n <name>Pastry</name>\n <bndbox>\n <xmin>0.08</xmin>\n <ymin>0.64</ymin>\n <xmax>0.17</xmax>\n <ymax>0.69</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/02fc5f4b8a661ded.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.45</ymin>\n <xmax>0.2</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/00ceead3fe54c7e7.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>House</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.21</ymin>\n <xmax>0.97</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Ladder</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.32</ymin>\n <xmax>0.6</xmax>\n <ymax>0.99</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/03a0722740bdc6d2.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Bronze sculpture</name>\n <bndbox>\n <xmin>0.32</xmin>\n <ymin>0.07</ymin>\n <xmax>0.66</xmax>\n <ymax>0.9</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human body</name>\n <bndbox>\n <xmin>0.36</xmin>\n <ymin>0.1</ymin>\n <xmax>0.67</xmax>\n <ymax>0.89</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tree</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0</ymin>\n <xmax>0.27</xmax>\n <ymax>0.58</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tree</name>\n <bndbox>\n <xmin>0.15</xmin>\n <ymin>0</ymin>\n <xmax>0.99</xmax>\n <ymax>0.66</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human arm</name>\n <bndbox>\n <xmin>0.41</xmin>\n <ymin>0.21</ymin>\n <xmax>0.65</xmax>\n <ymax>0.37</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/067b422f0eebb398.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Deer</name>\n <bndbox>\n <xmin>0.32</xmin>\n <ymin>0.45</ymin>\n <xmax>0.9</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/040a06c88281af41.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.45</xmin>\n <ymin>0.58</ymin>\n <xmax>0.5</xmax>\n <ymax>0.79</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tree</name>\n <bndbox>\n <xmin>0.81</xmin>\n <ymin>0</ymin>\n <xmax>0.98</xmax>\n <ymax>0.94</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tree</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0</ymin>\n <xmax>0.15</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tree</name>\n <bndbox>\n <xmin>0.31</xmin>\n <ymin>0</ymin>\n <xmax>0.89</xmax>\n <ymax>0.94</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tree</name>\n <bndbox>\n <xmin>0.37</xmin>\n <ymin>0</ymin>\n <xmax>0.41</xmax>\n <ymax>0.13</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/07aba4c8f2083a39.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Hat</name>\n <bndbox>\n <xmin>0.15</xmin>\n <ymin>0.01</ymin>\n <xmax>0.34</xmax>\n <ymax>0.16</ymax>\n </bndbox>\n </object>\n <object>\n <name>Hat</name>\n <bndbox>\n <xmin>0.33</xmin>\n <ymin>0.09</ymin>\n <xmax>0.45</xmax>\n <ymax>0.17</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.41</xmin>\n <ymin>0.34</ymin>\n <xmax>0.45</xmax>\n <ymax>0.42</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.54</xmin>\n <ymin>0.2</ymin>\n <xmax>0.59</xmax>\n <ymax>0.28</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/08491157edf5f550.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Porch</name>\n <bndbox>\n <xmin>0.04</xmin>\n <ymin>0.62</ymin>\n <xmax>0.11</xmax>\n <ymax>0.67</ymax>\n </bndbox>\n </object>\n <object>\n <name>Porch</name>\n <bndbox>\n <xmin>0.14</xmin>\n <ymin>0.66</ymin>\n <xmax>0.9</xmax>\n <ymax>0.81</ymax>\n </bndbox>\n </object>\n <object>\n <name>Porch</name>\n <bndbox>\n <xmin>0.37</xmin>\n <ymin>0.04</ymin>\n <xmax>0.69</xmax>\n <ymax>0.2</ymax>\n </bndbox>\n </object>\n <object>\n <name>Porch</name>\n <bndbox>\n <xmin>0.37</xmin>\n <ymin>0.23</ymin>\n <xmax>0.72</xmax>\n <ymax>0.33</ymax>\n </bndbox>\n </object>\n <object>\n <name>Porch</name>\n <bndbox>\n <xmin>0.74</xmin>\n <ymin>0.15</ymin>\n <xmax>0.93</xmax>\n <ymax>0.22</ymax>\n </bndbox>\n </object>\n <object>\n <name>Porch</name>\n <bndbox>\n <xmin>0.75</xmin>\n <ymin>0.27</ymin>\n <xmax>0.95</xmax>\n <ymax>0.29</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.25</xmin>\n <ymin>0.19</ymin>\n <xmax>0.36</xmax>\n <ymax>0.3</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.26</xmin>\n <ymin>0.35</ymin>\n <xmax>0.35</xmax>\n <ymax>0.4</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/07737659a761bdf2.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Box</name>\n <bndbox>\n <xmin>0.17</xmin>\n <ymin>0.13</ymin>\n <xmax>0.87</xmax>\n <ymax>0.71</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0166a2428257bb07.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Insect</name>\n <bndbox>\n <xmin>0.26</xmin>\n <ymin>0.48</ymin>\n <xmax>0.42</xmax>\n <ymax>0.59</ymax>\n </bndbox>\n </object>\n <object>\n <name>Common sunflower</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.31</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/06ed42f30446e8c4.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Human body</name>\n <bndbox>\n <xmin>0.57</xmin>\n <ymin>0.25</ymin>\n <xmax>0.83</xmax>\n <ymax>0.93</ymax>\n </bndbox>\n </object>\n <object>\n <name>Woman</name>\n <bndbox>\n <xmin>0.56</xmin>\n <ymin>0.17</ymin>\n <xmax>0.85</xmax>\n <ymax>0.91</ymax>\n </bndbox>\n </object>\n <object>\n <name>Mammal</name>\n <bndbox>\n <xmin>0.18</xmin>\n <ymin>0.18</ymin>\n <xmax>0.57</xmax>\n <ymax>0.92</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human body</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.22</ymin>\n <xmax>0.11</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human body</name>\n <bndbox>\n <xmin>0.23</xmin>\n <ymin>0.15</ymin>\n <xmax>0.52</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human body</name>\n <bndbox>\n <xmin>0.76</xmin>\n <ymin>0.29</ymin>\n <xmax>1</xmax>\n <ymax>0.91</ymax>\n </bndbox>\n </object>\n <object>\n <name>Mammal</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.21</ymin>\n <xmax>0.11</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Mammal</name>\n <bndbox>\n <xmin>0.56</xmin>\n <ymin>0.17</ymin>\n <xmax>0.84</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Mammal</name>\n <bndbox>\n <xmin>0.92</xmin>\n <ymin>0.52</ymin>\n <xmax>1</xmax>\n <ymax>0.92</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.66</xmin>\n <ymin>0.23</ymin>\n <xmax>0.75</xmax>\n <ymax>0.42</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.77</xmin>\n <ymin>0.42</ymin>\n <xmax>0.79</xmax>\n <ymax>0.46</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.83</xmin>\n <ymin>0.44</ymin>\n <xmax>0.87</xmax>\n <ymax>0.52</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.89</xmin>\n <ymin>0.43</ymin>\n <xmax>0.92</xmax>\n <ymax>0.49</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/05d63d6f4b194a53.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Tree</name>\n <bndbox>\n <xmin>0.1</xmin>\n <ymin>0.76</ymin>\n <xmax>0.18</xmax>\n <ymax>0.96</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tree</name>\n <bndbox>\n <xmin>0.82</xmin>\n <ymin>0.41</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/083a1d58d115b469.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Drum</name>\n <bndbox>\n <xmin>0.45</xmin>\n <ymin>0.68</ymin>\n <xmax>0.9</xmax>\n <ymax>0.99</ymax>\n </bndbox>\n </object>\n <object>\n <name>Drum</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.82</ymin>\n <xmax>0.36</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Drum</name>\n <bndbox>\n <xmin>0.96</xmin>\n <ymin>0.79</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/00478305a81396ef.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Sports uniform</name>\n <bndbox>\n <xmin>0.22</xmin>\n <ymin>0.03</ymin>\n <xmax>0.8</xmax>\n <ymax>0.97</ymax>\n </bndbox>\n </object>\n <object>\n <name>Baseball glove</name>\n <bndbox>\n <xmin>0.2</xmin>\n <ymin>0.18</ymin>\n <xmax>0.41</xmax>\n <ymax>0.33</ymax>\n </bndbox>\n </object>\n <object>\n <name>Sports uniform</name>\n <bndbox>\n <xmin>0.45</xmin>\n <ymin>0.14</ymin>\n <xmax>0.79</xmax>\n <ymax>0.42</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/00f4d4258b8954b0.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Suit</name>\n <bndbox>\n <xmin>0.07</xmin>\n <ymin>0.4</ymin>\n <xmax>0.7</xmax>\n <ymax>0.98</ymax>\n </bndbox>\n </object>\n <object>\n <name>Man</name>\n <bndbox>\n <xmin>0.07</xmin>\n <ymin>0.07</ymin>\n <xmax>0.76</xmax>\n <ymax>0.98</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human beard</name>\n <bndbox>\n <xmin>0.41</xmin>\n <ymin>0.34</ymin>\n <xmax>0.65</xmax>\n <ymax>0.64</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tree</name>\n <bndbox>\n <xmin>0.04</xmin>\n <ymin>0.52</ymin>\n <xmax>0.1</xmax>\n <ymax>0.59</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tree</name>\n <bndbox>\n <xmin>0.13</xmin>\n <ymin>0.31</ymin>\n <xmax>0.35</xmax>\n <ymax>0.58</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tree</name>\n <bndbox>\n <xmin>0.71</xmin>\n <ymin>0.52</ymin>\n <xmax>1</xmax>\n <ymax>0.99</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.42</xmin>\n <ymin>0.11</ymin>\n <xmax>0.66</xmax>\n <ymax>0.61</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/022cb88b012490b2.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Footwear</name>\n <bndbox>\n <xmin>0.5</xmin>\n <ymin>0.72</ymin>\n <xmax>0.56</xmax>\n <ymax>0.78</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.15</xmin>\n <ymin>0.37</ymin>\n <xmax>0.2</xmax>\n <ymax>0.43</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.23</xmin>\n <ymin>0.41</ymin>\n <xmax>0.27</xmax>\n <ymax>0.46</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.27</xmin>\n <ymin>0.36</ymin>\n <xmax>0.3</xmax>\n <ymax>0.42</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.38</xmin>\n <ymin>0.31</ymin>\n <xmax>0.4</xmax>\n <ymax>0.35</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.42</xmin>\n <ymin>0.38</ymin>\n <xmax>0.46</xmax>\n <ymax>0.43</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.48</xmin>\n <ymin>0.44</ymin>\n <xmax>0.52</xmax>\n <ymax>0.51</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.56</xmin>\n <ymin>0.38</ymin>\n <xmax>0.6</xmax>\n <ymax>0.45</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.66</xmin>\n <ymin>0.44</ymin>\n <xmax>0.7</xmax>\n <ymax>0.5</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.7</xmin>\n <ymin>0.41</ymin>\n <xmax>0.75</xmax>\n <ymax>0.49</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.83</xmin>\n <ymin>0.3</ymin>\n <xmax>0.87</xmax>\n <ymax>0.39</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.87</xmin>\n <ymin>0.45</ymin>\n <xmax>0.91</xmax>\n <ymax>0.53</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/004510be3d49b07c.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Human hair</name>\n <bndbox>\n <xmin>0.78</xmin>\n <ymin>0</ymin>\n <xmax>0.99</xmax>\n <ymax>0.36</ymax>\n </bndbox>\n </object>\n <object>\n <name>Footwear</name>\n <bndbox>\n <xmin>0.68</xmin>\n <ymin>0.95</ymin>\n <xmax>0.75</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Footwear</name>\n <bndbox>\n <xmin>0.84</xmin>\n <ymin>0.88</ymin>\n <xmax>0.9</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/072efa3a20cdda52.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Book</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0</ymin>\n <xmax>0.16</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/046effd1bf61c485.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Human hair</name>\n <bndbox>\n <xmin>0.39</xmin>\n <ymin>0.73</ymin>\n <xmax>0.51</xmax>\n <ymax>0.95</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human head</name>\n <bndbox>\n <xmin>0.38</xmin>\n <ymin>0.74</ymin>\n <xmax>0.52</xmax>\n <ymax>0.94</ymax>\n </bndbox>\n </object>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.56</xmin>\n <ymin>0.58</ymin>\n <xmax>0.68</xmax>\n <ymax>0.83</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human hair</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.8</ymin>\n <xmax>0.03</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human hair</name>\n <bndbox>\n <xmin>0.05</xmin>\n <ymin>0.84</ymin>\n <xmax>0.09</xmax>\n <ymax>0.9</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human hair</name>\n <bndbox>\n <xmin>0.15</xmin>\n <ymin>0.85</ymin>\n <xmax>0.2</xmax>\n <ymax>0.97</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human hair</name>\n <bndbox>\n <xmin>0.23</xmin>\n <ymin>0.8</ymin>\n <xmax>0.3</xmax>\n <ymax>0.9</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human hair</name>\n <bndbox>\n <xmin>0.51</xmin>\n <ymin>0.84</ymin>\n <xmax>0.58</xmax>\n <ymax>0.95</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human hair</name>\n <bndbox>\n <xmin>0.58</xmin>\n <ymin>0.87</ymin>\n <xmax>0.63</xmax>\n <ymax>0.94</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human hair</name>\n <bndbox>\n <xmin>0.58</xmin>\n <ymin>0.54</ymin>\n <xmax>0.61</xmax>\n <ymax>0.59</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human hair</name>\n <bndbox>\n <xmin>0.63</xmin>\n <ymin>0.85</ymin>\n <xmax>0.67</xmax>\n <ymax>0.91</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human hair</name>\n <bndbox>\n <xmin>0.68</xmin>\n <ymin>0.82</ymin>\n <xmax>0.72</xmax>\n <ymax>0.89</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human hair</name>\n <bndbox>\n <xmin>0.83</xmin>\n <ymin>0.83</ymin>\n <xmax>0.86</xmax>\n <ymax>0.91</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human hair</name>\n <bndbox>\n <xmin>0.92</xmin>\n <ymin>0.84</ymin>\n <xmax>0.97</xmax>\n <ymax>0.92</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human head</name>\n <bndbox>\n <xmin>0.58</xmin>\n <ymin>0.54</ymin>\n <xmax>0.61</xmax>\n <ymax>0.59</ymax>\n </bndbox>\n </object>\n <object>\n <name>Man</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.79</ymin>\n <xmax>0.06</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Man</name>\n <bndbox>\n <xmin>0.05</xmin>\n <ymin>0.86</ymin>\n <xmax>0.13</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Man</name>\n <bndbox>\n <xmin>0.18</xmin>\n <ymin>0.81</ymin>\n <xmax>0.34</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Man</name>\n <bndbox>\n <xmin>0.19</xmin>\n <ymin>0.84</ymin>\n <xmax>0.25</xmax>\n <ymax>0.96</ymax>\n </bndbox>\n </object>\n <object>\n <name>Man</name>\n <bndbox>\n <xmin>0.32</xmin>\n <ymin>0.73</ymin>\n <xmax>0.62</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Man</name>\n <bndbox>\n <xmin>0.65</xmin>\n <ymin>0.82</ymin>\n <xmax>0.75</xmax>\n <ymax>0.99</ymax>\n </bndbox>\n </object>\n <object>\n <name>Man</name>\n <bndbox>\n <xmin>0.91</xmin>\n <ymin>0.83</ymin>\n <xmax>0.99</xmax>\n <ymax>0.99</ymax>\n </bndbox>\n </object>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.87</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/00568eb5189586cb.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Bronze sculpture</name>\n <bndbox>\n <xmin>0.1</xmin>\n <ymin>0.35</ymin>\n <xmax>0.75</xmax>\n <ymax>0.71</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/05884c7019412e3d.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Personal flotation device</name>\n <bndbox>\n <xmin>0.5</xmin>\n <ymin>0.48</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/07a23b3ae2fe599e.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Street light</name>\n <bndbox>\n <xmin>0.53</xmin>\n <ymin>0.25</ymin>\n <xmax>0.56</xmax>\n <ymax>0.51</ymax>\n </bndbox>\n </object>\n <object>\n <name>Man</name>\n <bndbox>\n <xmin>0.24</xmin>\n <ymin>0.47</ymin>\n <xmax>0.51</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Man</name>\n <bndbox>\n <xmin>0.54</xmin>\n <ymin>0.38</ymin>\n <xmax>0.77</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/02be2e7d50c57176.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Animal</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.46</ymin>\n <xmax>1</xmax>\n <ymax>0.95</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/02cc618135a60455.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Man</name>\n <bndbox>\n <xmin>0.02</xmin>\n <ymin>0.48</ymin>\n <xmax>0.38</xmax>\n <ymax>0.98</ymax>\n </bndbox>\n </object>\n <object>\n <name>Man</name>\n <bndbox>\n <xmin>0.42</xmin>\n <ymin>0.2</ymin>\n <xmax>0.63</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.15</xmin>\n <ymin>0.32</ymin>\n <xmax>0.22</xmax>\n <ymax>0.42</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.18</xmin>\n <ymin>0.54</ymin>\n <xmax>0.28</xmax>\n <ymax>0.74</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.51</xmin>\n <ymin>0.22</ymin>\n <xmax>0.57</xmax>\n <ymax>0.35</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.88</xmin>\n <ymin>0.76</ymin>\n <xmax>0.95</xmax>\n <ymax>0.91</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/04e803c2a114a6b0.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Vehicle registration plate</name>\n <bndbox>\n <xmin>0.43</xmin>\n <ymin>0.73</ymin>\n <xmax>0.59</xmax>\n <ymax>0.79</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/028841f03c50f1a6.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.67</xmin>\n <ymin>0.72</ymin>\n <xmax>0.69</xmax>\n <ymax>0.83</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/026d6c7837e5f40a.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.36</xmin>\n <ymin>0.34</ymin>\n <xmax>0.86</xmax>\n <ymax>0.79</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human hair</name>\n <bndbox>\n <xmin>0.42</xmin>\n <ymin>0.33</ymin>\n <xmax>0.84</xmax>\n <ymax>0.62</ymax>\n </bndbox>\n </object>\n <object>\n <name>Television</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.18</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human nose</name>\n <bndbox>\n <xmin>0.68</xmin>\n <ymin>0.47</ymin>\n <xmax>0.74</xmax>\n <ymax>0.55</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/06f69d3e96782274.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Giraffe</name>\n <bndbox>\n <xmin>0.11</xmin>\n <ymin>0.21</ymin>\n <xmax>0.91</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0171886e5bd1315a.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Billboard</name>\n <bndbox>\n <xmin>0.07</xmin>\n <ymin>0.15</ymin>\n <xmax>0.97</xmax>\n <ymax>0.82</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/08106dc54136f7d2.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Common fig</name>\n <bndbox>\n <xmin>0.19</xmin>\n <ymin>0.52</ymin>\n <xmax>0.37</xmax>\n <ymax>0.76</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/03801399b37fe3ac.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.36</xmin>\n <ymin>0.31</ymin>\n <xmax>0.4</xmax>\n <ymax>0.43</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.36</xmin>\n <ymin>0.57</ymin>\n <xmax>0.41</xmax>\n <ymax>0.69</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.48</xmin>\n <ymin>0.31</ymin>\n <xmax>0.53</xmax>\n <ymax>0.43</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.49</xmin>\n <ymin>0.57</ymin>\n <xmax>0.53</xmax>\n <ymax>0.69</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
[
{
"content": [
{
"path": null,
"text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n",
"type": "text"
}
],
"role": "system"
},
{
"content": [
{
"path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/07eebed2755ca555.jpg",
"text": null,
"type": "image"
}
],
"role": "user"
},
{
"content": [
{
"path": null,
"text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Mobile phone</name>\n <bndbox>\n <xmin>0.29</xmin>\n <ymin>0.16</ymin>\n <xmax>0.75</xmax>\n <ymax>0.99</ymax>\n </bndbox>\n </object>\n <object>\n <name>Mobile phone</name>\n <bndbox>\n <xmin>0.29</xmin>\n <ymin>0</ymin>\n <xmax>0.64</xmax>\n <ymax>0.12</ymax>\n </bndbox>\n </object>\n</annotation>",
"type": "text"
}
],
"role": "assistant"
}
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.