Real-ESRGAN aims at developing Practical Algorithms for General Image Restoration. We extend the powerful ESRGAN to a practical restoration application (namely, Real-ESRGAN), which is trained with pure synthetic data.
If Real-ESRGAN is helpful in your photos/projects, please help to ⭐ this repo. Thanks😊 Other recommended projects: ▶️ GFPGAN ▶️ BasicSR ▶️ facexlib
[Paper] [Project Page] [Demo] Xintao Wang, Liangbin Xie, Chao Dong, Ying Shan Applied Research Center (ARC), Tencent PCG Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences
We have provided a pretrained model (RealESRGAN_x4plus.pth) with upsampling X4. Note that RealESRGAN may still fail in some cases as the real-world degradations are really too complex. Moreover, it may not perform well on human faces, text, etc, which will be optimized later.
Real-ESRGAN will be a long-term supported project (in my current plan 😃). It will be continuously updated in my spare time.