wzhgba

/

opendwm-models

like 1

This PR improves the model card for the CVD-STORM model, described in CVD-STORM: Cross-View Video Diffusion with Spatial-Temporal Reconstruction Model for Autonomous Driving.

It adds:

• The pipeline_tag: text-to-3d to better categorize the model's capabilities on the Hub.
• The library_name: diffusers, as the model builds upon Stable Diffusion architectures, enabling an automated code snippet.
• Links to the official paper, project page, and the GitHub repository.
• The full abstract for comprehensive understanding.
• A sample usage code snippet from the official GitHub repository, demonstrating layout-conditioned LiDAR generation.

Please review and merge if these updates align with the repository's goals.