DiffMAC: Diffusion Manifold Hallucination Correction for High Generalization Blind Face Restoration

arXiv:2403.10098v1 Announce Type: new
Abstract: Blind face restoration (BFR) is a highly challenging problem due to the uncertainty of degradation patterns. Current methods have low generalization across photorealistic and heterogeneous domains. In this paper, we propose a Diffusion-Information-Diffusion (DID) framework to tackle diffusion manifold hallucination correction (DiffMAC), which achieves high-generalization face restoration in diverse degraded scenes and heterogeneous domains. Specifically, the first diffusion stage aligns the restored face with spatial feature embedding of the low-quality face based on AdaIN, which …

abstract arxiv blind cs.cv current diffusion domains face framework hallucination information low manifold paper patterns photorealistic type uncertainty