Real-ESRGAN

Real-SR improved with better discriminators and synthetically generated data

Released in: Real-ESRGAN: Training Real-World Blind Super-Resolution with Pure Synthetic Data

Contributor:

Summary

Though many attempts have been made in blind super-resolution to restore low-resolution images with unknown and complex degradations, they are still far from addressing general real-world degraded images. This work extends the powerful ESRGAN to a practical restoration application (namely, Real-ESRGAN), which is trained with pure synthetic data. Specifically, a high-order degradation modeling process is introduced to better simulate complex real-world degradations. The authors also consider the common ringing and overshoot artifacts in the synthesis process. In addition, they employ a U-Net discriminator with spectral normalization to increase discriminator capability and stabilize the training dynamics. Extensive comparisons have shown its superior visual performance than prior works on various real datasets. The paper also provides efficient implementations to synthesize training pairs on the fly.

2021

Year Released

Key Links & Stats

Real-ESRGAN

Real-ESRGAN: Training Real-World Blind Super-Resolution with Pure Synthetic Data

Real-ESRGAN: Training Real-World Blind Super-Resolution with Pure Synthetic Data

@InProceedings{wang2021realesrgan, author = {Xintao Wang and Liangbin Xie and Chao Dong and Ying Shan}, title = {Real-ESRGAN: Training Real-World Blind Super-Resolution with Pure Synthetic Data}, booktitle = {International Conference on Computer Vision Workshops (ICCVW)}, date = {2021} }

ML Tasks

  1. Super-Resolution

ML Platform

  1. Pytorch

Modalities

  1. Still Image

Verticals

  1. General

CG Platform

  1. Not Applicable

Related organizations

Tencent

Shenzhen Institutes of Advanced Technology