Fanghua-Yu/SUPIR

[CVPR 2025] FaithDiff for Classic Film Rejuvenation, Old Photo Revival, Social Media Restoration, Image Enhancement and AIGC Enhancement.

[ECCV2024] Pixel-Aware Stable Diffusion for Realistic Image Super-Resolution and Personalized Stylization

Real-ESRGAN aims at developing Practical Algorithms for General Image/Video Restoration.

Official code repository for "PixRestore: Unified Image Restoration via Pixel Diffusion Transformer".

[NeurIPS 24] PromptFix: You Prompt and We Fix the Photo

My self-trained AI models for photorealistic image upscaling and restoration. Built on ESRGAN, SAFMN, RealPLKSR and other architectures and trained on a manually curated high-end dataset (11,750+ real-world HD images). Models excel at restoring authentic textures and objects (skin, fabric, grit) and handle various types of degradation.

Restore Anything Pipeline (RAP): Segment Anything Meets Image Restoration

All my self trained & released AI upscaling models. After gathering and applying over 600 different upscaling models, I learned how to train my own models, and these are the results.

[ECCV 2024] codes of DiffBIR: Towards Blind Image Restoration with Generative Diffusion Prior