Reproduce “Scale Guided Hypernetwork for Blind Super-Resolution Image Quality Assessment” 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
Jun Fu
SGH is a blind super-resolution image-quality assessor that uses the width ratio of the super-resolved and low-resolution images to adapt its scoring rule. A hypernetwork generates weights and biases of quality-prediction layers from this scale, rather than concatenating scale with image features. On CVIU, SGH improved average SRCC by 0.1660 versus 0.1208 for fusion across five models, though results varied.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.