AI RESEARCH
Beyond Ground-Truth: Leveraging Image Quality Priors for Real-World Image Restoration
arXiv CS.CV
•
ArXi:2603.29773v1 Announce Type: new Real-world image restoration aims to re high-quality (HQ) images from degraded low-quality (LQ) inputs captured under uncontrolled conditions. Existing methods typically depend on ground-truth (GT) supervision, assuming that GT provides perfect reference quality. However, GT can still contain images with inconsistent perceptual fidelity, causing models to converge to the average quality level of the