The Perception-Robustness Tradeoff in Deterministic Image Restoration

arXiv:2311.09253v2 Announce Type: replace-cross
Abstract: We study the behavior of deterministic methods for solving inverse problems in imaging. These methods are commonly designed to achieve two goals: (1) attaining high perceptual quality, and (2) generating reconstructions that are consistent with the measurements. We provide a rigorous proof that the better a predictor satisfies these two requirements, the larger its Lipschitz constant must be, regardless of the nature of the degradation involved. In particular, to approach perfect perceptual quality and perfect …

abstract arxiv behavior consistent cs.cv cs.lg eess.iv eess.sp image image restoration imaging perception quality restoration robustness study type