SRFlow: Learning the Super-Resolution Space with Normalizing Flow

Code

• github.com/andreas128/SRFlow
  OfficialIn paperpytorch★ 854
• github.com/liyuantsao/flowsr-lp
  pytorch★ 85
• github.com/liyuantsao/BFSR
  pytorch★ 85
• github.com/Zhangyanbo/iResNetLab
  pytorch★ 71
• github.com/seungho-snu/fxsr
  pytorch★ 15
• github.com/friedmanroy/hi-generation
  pytorch★ 3
• github.com/2023-MindSpore-1/ms-code-222/tree/main/SRFlow
  mindspore★ 0
• github.com/andreas128/NTIRE21_Learning_SR_Space
  pytorch★ 0

Abstract

Super-resolution is an ill-posed problem, since it allows for multiple predictions for a given low-resolution image. This fundamental fact is largely ignored by state-of-the-art deep learning based approaches. These methods instead train a deterministic mapping using combinations of reconstruction and adversarial losses. In this work, we therefore propose SRFlow: a normalizing flow based super-resolution method capable of learning the conditional distribution of the output given the low-resolution input. Our model is trained in a principled manner using a single loss, namely the negative log-likelihood. SRFlow therefore directly accounts for the ill-posed nature of the problem, and learns to predict diverse photo-realistic high-resolution images. Moreover, we utilize the strong image posterior learned by SRFlow to design flexible image manipulation techniques, capable of enhancing super-resolved images by, e.g., transferring content from other images. We perform extensive experiments on faces, as well as on super-resolution in general. SRFlow outperforms state-of-the-art GAN-based approaches in terms of both PSNR and perceptual quality metrics, while allowing for diversity through the exploration of the space of super-resolved solutions.

Tasks

Benchmark Results

Reproductions