STAR
First spatiotemporal-enhanced video super-resolution combining T2V diffusion models
The first spatiotemporal-enhanced video super-resolution combining T2V diffusion models.
AI video generation, digital humans, and video restoration.
Key features
- Real-world video super-resolution using text-to-video models
- Spatial-temporal augmentation approach
- Two base model versions: I2VGen-XL and CogVideoX-5B
- Light and heavy degradation variants; Colab and Hugging Face demos
Good for
- Upscaling low-quality videos
- Restoring videos downloaded from the web
Stack & tags
Pythonvideo-super-resolutionvideo-upscalingopen-weightst2vi2vgencogvideoxrestoration
Quick start
bash video_super_resolution/scripts/inference_sr.shFrom the project README on GitHub.
Similar open-source projects
Hyperframes
★ 55,468Write HTML. Render video. Built for agents — HeyGen's open-source video framework
Hypit
★ 18,649Clone any viral video with AI agents — swap the face, the words, the B-roll, ship 100 variants