Skip to content

Implement a closer paper-faithful module #1

Description

@StaryMoon

The v0.1.0 release is a compact toy scaffold. This issue tracks replacing the placeholder module with a closer implementation of: Diffusion Renderer: Neural Inverse and Forward Rendering with Video Diffusion Models

Expected work:

  • map paper equations or architecture blocks to code
  • keep tensor shapes documented
  • add a smoke test
  • update README without claiming reproduced benchmarks

Metadata

Metadata

Assignees

No one assigned

    Labels

    help wantedExtra attention is neededimplementationPaper-to-code implementation gap

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions