Effective Adapter for Face Recognition in the Wild

arXiv:2312.01734v2 Announce Type: replace
Abstract: In this paper, we tackle the challenge of face recognition in the wild, where images often suffer from low quality and real-world distortions. Traditional heuristic approaches-either training models directly on these degraded images or their enhanced counterparts using face restoration techniques-have proven ineffective, primarily due to the degradation of facial features and the discrepancy in image domains. To overcome these issues, we propose an effective adapter for augmenting existing face recognition models trained on high-quality …

abstract adapter arxiv challenge cs.cv face face recognition images low paper quality recognition training training models type world