Dataset · FelixYuan-YF
Large video dataset with spatial annotations
FelixYuan-YF/SpatialVID-HQ
A large-scale dataset featuring videos and spatial annotations for 3D and video generation tasks.
What's inside
The dataset from researchers at Nanjing University and the Chinese Academy of Sciences (CVPR 2026 paper) contains between 100,000 and 1 million records. The metadata specifies English tabular and text modalities associated with a large-scale video corpus with spatial annotations.
What you can do with it
- Training text-to-video and image-to-video generation models
- Training 3D reconstruction and three-dimensional scene generation algorithms (text-to-3D, image-to-3D)
- Tasks
- Text → video, Text → 3d, Image → 3d, Image → video, Other
- Languages
- English
- Size
- 100K–1M rows
- Downloads this month
- 5.6K
- Likes
- 35
- Access
- on request on Hugging Face
How to load it
from datasets import load_dataset
ds = load_dataset("FelixYuan-YF/SpatialVID-HQ")