Yuheng02

/

OmniRoam

like 4

Hi! I'm opening this PR to improve the model card with relevant metadata and better discoverability.

Specifically, I have:

• Added the image-to-video pipeline tag to help users find the model.
• Specified the license as other (Adobe Research License).
• Added base_model: Wan-AI/Wan2.1-T2V-1.3B to show the lineage of the model.
• Maintained the existing documentation, including setup and inference instructions.

The model is associated with the paper: OmniRoam: World Wandering via Long-Horizon Panoramic Video Generation.