ChenHsing/Awesome-Video-Diffusion-Models

Comprehensive curated resource covering diffusion-based approaches for video generation and editing tasks, with categorized sections spanning text-to-video synthesis, pose/instruction/sound-guided generation, video completion, and editing methods. Organizes research across training-based and training-free paradigms, plus foundational models and toolboxes like Stable Video Diffusion, AnimateDiff, and Open-Sora implementations. Serves as a taxonomy of diffusion architectures—from U-Net and Transformer variants to latent diffusion approaches—enabling researchers to identify methodological patterns across video generation and manipulation applications.

2,282 stars. Actively maintained with 5 commits in the last 30 days.