Reproduce “Learning a Single Convolutional Super-Resolution Network for Multiple Degradations” 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

Kai Zhang, Wangmeng Zuo, Lei Zhang

SRMD performs single-image super-resolution by giving a CNN the low-resolution image and maps describing blur and noise. It encodes the blur kernel with principal component analysis and combines degradation maps with image pixels. On Set5 at ×3, SRMD scores 25.82 dB versus 25.19 dB for DnCNN+SRMDNF under specified blur, bicubic downsampling, and noise; degradation is known, not estimated.

Summary by alphaXiv.

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