Dataset · bones-studio
Annotated human motions for robots
bones-studio/seed
142,220 human motion animations with text descriptions and annotations for humanoid robotics.
What's inside
BONES-SEED is an open dataset of 142,220 annotated human motion animations. Motion capture data is provided in SOMA and Unitree G1 formats, featuring natural language descriptions, temporal segmentation, and skeleton metadata. The description language is English.
What you can do with it
- Train humanoid robot control policies
- Generate motion from text descriptions
- Evaluate video-text-to-text models on motion descriptions
- Tasks
- Robotics, Text → video, Video text → text
- Languages
- English
- Size
- 100K–1M rows
- Downloads this month
- 17.1K
- Likes
- 514
- Access
- on request on Hugging Face
How to load it
from datasets import load_dataset
ds = load_dataset("bones-studio/seed")