Text-to-Video

Enhance model card: Add pipeline tag, paper, project, code links, and usage

#1
by nielsr HF Staff - opened

This PR significantly enhances the model card for CineScale, improving its discoverability and providing comprehensive documentation for users.

Key additions include:

  • Pipeline Tag: Added pipeline_tag: text-to-video to ensure the model appears in the relevant category on the Hub, aligning with its capabilities in high-resolution video generation.
  • Paper Link: Linked to the official paper, CineScale: Free Lunch in High-Resolution Cinematic Visual Generation.
  • External Links: Included clear links to the project page (https://eyeline-labs.github.io/CineScale/) and the GitHub repository (https://github.com/Eyeline-Labs/CineScale) for easy access to code and further details.
  • Model Overview: Added a table detailing the available CineScale models, their tuning resolutions, and capabilities.
  • Sample Usage: Provided a "Quick Start" section with a code snippet directly from the GitHub README, guiding users on how to initiate video generation, with a clear reference to the GitHub repository for more advanced examples and setup.
  • Citation: Included the BibTeX citation for proper academic attribution.

These changes make the model card more informative and user-friendly.

MoonQiu changed pull request status to merged
Eyeline Labs org

Thanks for it.

Sign up or log in to comment