AI RESEARCH
Dual-Path Hyperprior Informed Deep Unfolding Network for Image Compressive Sensing
arXiv CS.CV
•
ArXi:2605.09566v1 Announce Type: new Recent Deep Unfolding Networks (DUNs) have significantly advanced Compressive Sensing (CS) by integrating iterative optimization with deep networks. However, existing DUNs still suffer from two challenges: 1) Reliance on a single measurement stream, which limits effective information interaction across distinct measurement subsets. 2) Uniform processing of all image regions, which overlooks varying reconstruction difficulties induced by diverse textures.