Reproduce “Learning the Degradation Distribution for Blind Image Super-Resolution” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Zhengxiong Luo, Yan Huang, Shang Li, Liang Wang, Tieniu Tan

The authors present a blind image super-resolution system that learns from synthetic image pairs despite unknown degradation. Its probabilistic degradation model samples random inputs to generate blur kernels and noise, producing varied low-resolution training images from high-resolution images. On NTIRE2018 Track 4, sampling both factors improves PSNR, SSIM, and LPIPS over fixed inputs; this evidence is limited to that experiment.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.