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

Abstract

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. In this work, we extend 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. We also consider the common ringing and overshoot artifacts in the synthesis process. In addition, we 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. We also provide efficient implementations to synthesize training pairs on the fly.

Tech Report. Training/testing codes and executable files are in https://github.com/xinntao/Real-ESRGAN

Cited by in corpus (6)

• SwinIR: Image Restoration Using Swin Transformer
• A Systematic Survey of Deep Learning-based Single-Image Super-Resolution
• ERQA: Edge-Restoration Quality Assessment for Video Super-Resolution
• FreqNet: A Frequency-domain Image Super-Resolution Network with Dicrete Cosine Transform
• Finding Discriminative Filters for Specific Degradations in Blind Super-Resolution
• High Dynamic Range Image Reconstruction via Deep Explicit Polynomial Curve Estimation