Rethink Arbitrary Style Transfer with Transformer and Contrastive Learning

arXiv:2404.13584v1 Announce Type: cross
Abstract: Arbitrary style transfer holds widespread attention in research and boasts numerous practical applications. The existing methods, which either employ cross-attention to incorporate deep style attributes into content attributes or use adaptive normalization to adjust content features, fail to generate high-quality stylized images. In this paper, we introduce an innovative technique to improve the quality of stylized images. Firstly, we propose Style Consistency Instance Normalization (SCIN), a method to refine the alignment between content and style …

abstract applications arxiv attention cs.cv cs.lg features generate images normalization paper practical quality research style style transfer transfer transformer type