Reproduce “Disentangle Before Anonymize: A Two-stage Framework for Attribute-preserved and Occlusion-robust De-identification” 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
Mingrui Zhu, Dongxin Chen, Xin Wei, Nannan Wang, Xinbo Gao
DBAF separates identity from attributes before anonymizing face images, aiming to conceal who is pictured while retaining features such as expression, hairstyle, and background and allowing keyed recovery. Its staged design changes identity using a key, then recombines retained attributes. On CelebA-HQ, anonymized-image FID is 9.61 versus 16.375 for a single-stage ablation, supporting staging in this setup.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.