EricGuo5513/momask-codes
A CVPR 2024 generative model that synthesizes 3D human motions from text descriptions using masked modeling.
Not currently ranked — collecting fresh signals.
star history
MoMask implements a masked modeling approach for generating 3D skeletal human motions from textual descriptions. The model learns to reconstruct motion sequences from masked tokens, enabling controllable motion synthesis. It provides training scripts, evaluation tools, and a WebUI demo on HuggingFace Spaces for generating animation-ready BVH files from text prompts.
Frequently asked
- What is EricGuo5513/momask-codes?
- A CVPR 2024 generative model that synthesizes 3D human motions from text descriptions using masked modeling.
- Is momask-codes open source?
- Yes — EricGuo5513/momask-codes is open source, released under the MIT license.
- What language is momask-codes written in?
- EricGuo5513/momask-codes is primarily written in Python.
- How popular is momask-codes?
- EricGuo5513/momask-codes has 1.3k stars on GitHub.
- Where can I find momask-codes?
- EricGuo5513/momask-codes is on GitHub at https://github.com/EricGuo5513/momask-codes.