Add comprehensive model card for MAGREF with metadata and usage

#1
by nielsr HF Staff - opened

This PR significantly enhances the model card for MAGREF: Masked Guidance for Any-Reference Video Generation with Subject Disentanglement.

Key additions include:

  • Metadata tags: pipeline_tag: image-to-video, library_name: diffusers, and license: mit.
  • Direct links to the paper, project page, and GitHub repository.
  • A detailed description of the model, including its abstract, features, and recent news, extracted from the paper and GitHub README.
  • A demo video from the GitHub README to showcase model capabilities.
  • Comprehensive installation instructions and a "Quick Start" guide for inference (single and multi-GPU), including explicit code snippets from the GitHub README.
  • Acknowledgements, ethics concerns, and citation information.

These improvements will make the model more discoverable and easier to use for the community, providing all necessary information in one place. Please review and merge if everything looks good!

Ready to merge
This branch is ready to get merged automatically.

Sign up or log in to comment