Reproduce “BasicVSR++: Improving Video Super-Resolution with Enhanced Propagation and Alignment” 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

Kelvin C.K. Chan, Shangchen Zhou, Xiangyu Xu, Chen Change Loy

BasicVSR++ improves video super-resolution by refining recurrent propagation and motion-guided feature alignment. It uses bidirectional second-order grid propagation and optical-flow-guided deformable sampling. On REDS4 at 4× downsampling, its lighter variant reaches 32.24 dB PSNR versus BasicVSR’s 31.42 dB, with similar parameter counts and comparable reported runtimes; this supports improved fidelity in this tested comparison, not faster runtime.

Summary by alphaXiv.

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