Improve model card: Add metadata, paper & code links, abstract, and sample usage
#1
by
nielsr
HF Staff
- opened
This PR significantly enhances the model card for SyncHuman: Synchronizing 2D and 3D Generative Models for Single-view Human Reconstruction by:
- Adding the
pipeline_tag: image-to-3dto accurately categorize the model on the Hub and improve discoverability. - Specifying
library_name: diffusersbased on explicit references in the model's configuration files and dependencies listed in the GitHub README, which enables an automated "how to use" widget. - Including a direct link to the official Hugging Face paper page: https://huggingface.co/papers/2510.07723.
- Adding a link to the GitHub repository: https://github.com/IGL-HKUST/SyncHuman.
- Incorporating the paper's abstract to provide a comprehensive overview of the model.
- Adding a detailed "Sample Usage" section, including environment setup and inference instructions with code snippets, directly extracted from the official GitHub README.
The existing license: mit and base_model: - microsoft/TRELLIS-image-large metadata have been retained.
These improvements will make the model more informative and user-friendly for the Hugging Face community.