integrate_Cap3D_captions

[tiangeluo]

This PR integrates Cap3D captions into Objaverse annotations along with point-clouds download link.

objaverse.load_uids() will first download a cap3d_captions.json.gz file locally and then assign captions to each inquired UID. For pointclouds, user should follow the download link to download the zip files with point clouds stored as .ply and .pt.

Specifically, the Cap3D caption and the point-cloud download link are displayed in the last two lines. One example is the below.

import objaverse
uids = objaverse.load_uids()
annotations = objaverse.load_annotations(uids)

annotations["ysqP6VH2x99bODrQRbcYdQjSLjV"]

{'uri': 'https://api.sketchfab.com/v3/models/ysqP6VH2x99bODrQRbcYdQjSLjV',
 'uid': 'ysqP6VH2x99bODrQRbcYdQjSLjV',
 'name': 'Tiko',
 'staffpickedAt': '2013-06-19T08:56:52',
 'viewCount': 937,
 'likeCount': 27,
 'animationCount': 0,
 'viewerUrl': 'https://sketchfab.com/3d-models/ysqP6VH2x99bODrQRbcYdQjSLjV',
 'embedUrl': 'https://sketchfab.com/models/ysqP6VH2x99bODrQRbcYdQjSLjV/embed',
 'commentCount': 4,
 'isDownloadable': True,
 'publishedAt': '2013-06-17T05:01:56',
 'tags': [{'name': '3dcoat',
   'slug': '3dcoat',
   'uri': 'https://api.sketchfab.com/v3/tags/3dcoat'},
  {'name': 'sketchfast5',
   'slug': 'sketchfast5',
   'uri': 'https://api.sketchfab.com/v3/tags/sketchfast5'},
  {'name': 'staffpicks',
   'slug': 'staffpicks',
   'uri': 'https://api.sketchfab.com/v3/tags/staffpicks'},
  {'name': 'blender',
   'slug': 'blender',
   'uri': 'https://api.sketchfab.com/v3/tags/blender'},
  {'name': 'art',
   'slug': 'art',
   'uri': 'https://api.sketchfab.com/v3/tags/art'}],
 'categories': [],
 'thumbnails': {'images': [{'uid': '962ec9c15cbf40d48e531ae9ea37245c',
    'size': 157076,
    'width': 1920,
    'url': 'https://media.sketchfab.com/models/ysqP6VH2x99bODrQRbcYdQjSLjV/thumbnails/537654ca708642c29b533e689e5f8b5b/b8c2ee6aa6e04bf6b4066b76e0e1c477.jpeg',
    'height': 1080},
   {'uid': 'b9f16c31871049bea4937d291759310c',
    'size': 41138,
    'width': 1024,
    'url': 'https://media.sketchfab.com/models/ysqP6VH2x99bODrQRbcYdQjSLjV/thumbnails/537654ca708642c29b533e689e5f8b5b/41428fa34289463fa66054479d366905.jpeg',
    'height': 576},
   {'uid': 'f69b19b325b34fe18b1814585aa9c25e',
    'size': 22768,
    'width': 720,
    'url': 'https://media.sketchfab.com/models/ysqP6VH2x99bODrQRbcYdQjSLjV/thumbnails/537654ca708642c29b533e689e5f8b5b/bd83c4a46304443c94772f419936d507.jpeg',
    'height': 405},
   {'uid': 'cdc9a424136440d1abe8fc982e05fc9e',
    'size': 4606,
    'width': 256,
    'url': 'https://media.sketchfab.com/models/ysqP6VH2x99bODrQRbcYdQjSLjV/thumbnails/537654ca708642c29b533e689e5f8b5b/bd122ebde94748e19a947cb6d3e75b37.jpeg',
    'height': 144},
   {'uid': '744290d632c849e4a2e8e1d8f1f9bd53',
    'size': 820,
    'width': 64,
    'url': 'https://media.sketchfab.com/models/ysqP6VH2x99bODrQRbcYdQjSLjV/thumbnails/537654ca708642c29b533e689e5f8b5b/bae26559d66e419d8af802d1a1cd65cb.jpeg',
    'height': 36}]},
 'user': {'uid': 'ee628401f8414f15a8b72ea5586f7822',
  'username': 'zafio',
  'displayName': 'Zafio',
  'profileUrl': 'https://sketchfab.com/zafio',
  'account': 'basic',
  'avatar': {'uri': 'https://api.sketchfab.com/v3/avatars/b19596b7e6534d69bed828007e99f579',
   'images': [{'size': 641,
     'width': 32,
     'url': 'https://media.sketchfab.com/avatars/b19596b7e6534d69bed828007e99f579/efe9788ae68f443881ff834b5934be86.jpeg',
     'height': 32},
    {'size': 953,
     'width': 48,
     'url': 'https://media.sketchfab.com/avatars/b19596b7e6534d69bed828007e99f579/6e9e9358d62d48a8a42268e182f754ac.jpeg',
     'height': 48},
    {'size': 2039,
     'width': 90,
     'url': 'https://media.sketchfab.com/avatars/b19596b7e6534d69bed828007e99f579/9dcc185f3f914074a0b8632a79fe193c.jpeg',
     'height': 90},
    {'size': 2322,
     'width': 100,
     'url': 'https://media.sketchfab.com/avatars/b19596b7e6534d69bed828007e99f579/5ea780bebf384360bedd117c3adc68ec.jpeg',
     'height': 100}]},
  'uri': 'https://api.sketchfab.com/v3/users/ee628401f8414f15a8b72ea5586f7822'},
 'description': '',
 'faceCount': 49862,
 'createdAt': '2013-06-17T05:01:56',
 'vertexCount': 25242,
 'isAgeRestricted': False,
 'archives': {'glb': {'textureCount': 5,
   'size': 11189196,
   'type': 'glb',
   'textureMaxResolution': 2048,
   'faceCount': 54470,
   'vertexCount': 30300},
  'gltf': {'textureCount': 5,
   'size': 10025782,
   'type': 'gltf',
   'textureMaxResolution': 2048,
   'faceCount': 54470,
   'vertexCount': 30300},
  'source': {'textureCount': None,
   'size': 8120079,
   'type': 'source',
   'textureMaxResolution': None,
   'faceCount': None,
   'vertexCount': None},
  'usdz': {'textureCount': None,
   'size': 6374448,
   'type': 'usdz',
   'textureMaxResolution': None,
   'faceCount': None,
   'vertexCount': None}},
 'license': 'by',
 'cap3d': {'caption': 'a small purple toy monster.',
  'pointcloud_download_url': 'https://huggingface.co/datasets/tiange/Cap3D/tree/main/PointCloud_zips',
  'paper': 'Scalable 3D Captioning with Pretrained Models'}}