Visual Genome 数据集 Visual Genome 数据集格式
<h2>1. Images</h2>
- File image part1, image part2
全部 jpg 格式的图片
</li> </ul> <h2>2. Image meta data</h2>IMAGE_ID.jpg,- File image_data.json.zip
全部图片的 meta data，格式：
如：
</li> </ul> <h2>3. Region Descriptions</h2>[... { "image_id": 2412112, "url": "https://cs.stanford.edu/people/rak248/VG_100K/2370463.jpg", "width": 500, "height": 281, "coco_id": 547168, "flickr_id": 8505158818 } ...]- File region_descriptions.json.zip
全部的 region descriptions.
如：
</li> </ul> <h2>4. Question Answers (QAs)</h2>
[... { "image_id": 2407890, "regions": [... { "region_id": 1353, "x": 117, "y": 79, "width": 249, "height": 107, "phrase": "a cat sitting on a table.", "synsets": [... { "synset_name": "cat.n.01", "entity_name": "cat", "entity_idx_start": 2, "entity_idx_end": 5 }, ...] }, { "region_id": 1354, "x": 116, "y": 29, "width": 239, "height": 135, "phrase": "a white cat with a tan tail and face markings", "synsets": [... ...] }, ...] }, { "image_id": 2407890, "regions": [... ...] }, ...]- File question_answers.json.zip
全部的问题与答案(QAs).
[... { "image_id": 2317993, "qas": [... { "qa_id": 912402, "question": "Where are the clouds?", "answer": "sky", "question_synsets": [... { "synset_name": "cloud.n.01", "entity_name": "cloud", "entity_idx_start": 14, "entity_idx_end": 20 }, ...], "answer_synsets": [... { "synset_name": "sky.n.01", "entity_name": "sky", "entity_idx_start": 0, "entity_idx_end": 3 }, ...] }, ...] }, ...]</p></li>
</ul><h2>5. Objects</h2>
- <p>File objects.json.zip
全部的 object 实例.
[... { "image_id": 2, "objects": [... { "object_id": 1023847, "x": 405, "y": 34, "w": 78, "h": 438, "name": "pole", "synsets": ["pole.n.01"] }, { "object_id": 1023836, "x": 239, "y": 347, "w": 136, "h": 126, "name": "car", "synsets": ["car.n.01"] }, ...] }, ...]</p></li>
</ul><h2>6. Attributes</h2>
- <p>File attributes.json.zip
数据集中全部的 attributes.
如：
[... { "image_id": 2, "attributes": [... { "object_id": 1023847, "x": 405, "y": 34, "w": 78, "h": 438, "name": "pole", "synsets": ["pole.n.01"], "attributes": ["brown"] }, { "object_id": 1023836, "x": 239, "y": 347, "w": 136, "h": 126, "name": "car", "synsets": ["car.n.01"], "attributes": ["red", "broken"] }, ...] }, ...]</p></li>
</ul><h2>7. Relationships</h2>
- <p>File relationships.json.zip
全部的 relationships.
如：
</li> </ul> <h2>8. Synset Name & Descriptions</h2>[... { "image_id": 2, "relationships": [... { "relationship_id": 15947, "predicate": "wears", "synsets": ["wear.v.01"], "subject": { "object_id": 1023838, "x": 324, "y": 320, "w": 142, "h": 255, "name": "man", "synsets": ["man.n.01"] }, "object": { "object_id": 5071, "x": 359, "y": 362, "w": 72, "h": 81, "name": "backpack", "synsets": ["backpack.n.01"] }, }, ...], } ...]- File synsets.json.zip
全部的同义词及其描述.
如：
[... { "synset_name": "phonograph_record.n.01", "synset_definition": "sound recording consisting of a disk with a continuous groove; used to reproduce music by rotating while a phonograph needle tracks in the groove", }, { "synset_name": "truck.n.01", "synset_definition": "an automotive vehicle suitable for hauling", } ...]</p></li>
</ul><h2>9. Region Graphs</h2>
- <p>File region_graphs.json.zip
全部的 region graphs.
如：
[... { "image_id": 2407890, "regions": [... { "region_id": 1353, "x": 117, "y": 79, "width": 249, "height": 107, "phrase": "a cat sitting on a table.", "synsets": [... { "synset_name": "cat.n.01", "entity_name": "cat", "entity_idx_start": 2, "entity_idx_end": 5 }, ...] "objects": [... { "object_id": 1023838, "x": 324, "y": 320, "w": 142, "h": 255, "name": "cat", "synsets": ["cat.n.01"] }, { "object_id": 5071, "x": 359, "y": 362, "w": 72, "h": 81, "name": "table", "synsets": ["table.n.01"] }, ...], "relationships": [... { "relationship_id": 15947, "predicate": "wears", "synsets": ["wear.v.01"], "subject_id": 1023838, "object_id": 5071, } ...] }, ...] }, ...]</p></li>
</ul><h2>10. Scene Graphs</h2>
- <p>File scene_graphs.json.zip
全部的 scene graphs.
如：
[... { "image_id": 2407890, "objects": [... { "object_id": 1023838, "x": 324, "y": 320, "w": 142, "h": 255, "name": "cat", "synsets": ["cat.n.01"] }, { "object_id": 5071, "x": 359, "y": 362, "w": 72, "h": 81, "name": "table", "synsets": ["table.n.01"] }, ...], "relationships": [... { "relationship_id": 15947, "predicate": "wears", "synsets": ["wear.v.01"], "subject_id": 1023838, "object_id": 5071, } ...] }, ...]</p></li>
</ul><h2>11. Mapping from region based QA to region descriptions</h2>
- <p>File qa_to_region_mapping.json.zip
将 QA 映射到对应的 region descriptions.
如：
{... QA_ID: REGION_DESCRIPTION_ID, "1885736": "2072251" ...}<p></p></li>
- <p>File qa_to_region_mapping.json.zip
将 QA 映射到对应的 region descriptions.
如：
- <p>File scene_graphs.json.zip
全部的 scene graphs.
- <p>File region_graphs.json.zip
全部的 region graphs.
- File synsets.json.zip
全部的同义词及其描述.
- <p>File relationships.json.zip
全部的 relationships.
- <p>File attributes.json.zip
数据集中全部的 attributes.
- <p>File objects.json.zip
全部的 object 实例.
- File question_answers.json.zip
全部的问题与答案(QAs).
- File region_descriptions.json.zip
全部的 region descriptions.
- File image_data.json.zip
全部图片的 meta data，格式：