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

TL;DR

Real-ESRGAN is a novel super-resolution method trained solely on synthetic data that effectively models complex real-world degradations, leading to superior visual restoration performance.

Contribution

It introduces a high-order degradation model and a U-Net discriminator with spectral normalization for stable training, advancing real-world blind super-resolution.

Findings

Outperforms prior methods on real datasets in visual quality

Uses synthetic data for training without real degraded images

Provides efficient on-the-fly training pair synthesis

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.