Reproduce “CusEnhancer: A Zero-Shot Scene and Controllability Enhancement Method for Photo Customization via ResInversion” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Maoye Ren, Praneetha Vaddamanu, Jianjin Xu, Fernando De la Torre Frade
CustomEnhancer combines an SDXL-generated scene, a face-swapped reference identity, and a personalized model’s diffusion process to produce customized photos. Its backward and forward flows integrate scene information with identity customization and support SDXL-compatible controls without controller retraining. In the reported PhotoMaker evaluation, it scores 60.0% for scene diversity and 67.3% for face similarity; these findings apply to that evaluation setup.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.