跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2604.11250 · v2

TEMPER: Reducing Attribute Leakage from Face Recognition Embeddings

首次提交 2026-04-13 18:01
版本更新 2026-11-17 02:21
雷达观测 2026-11-17 16:10
arXiv ↗PDF ↗

摘要

Face recognition embeddings encode soft-biometric attributes such as gender and ethnicity, which can be inferred from templates. This paper proposes TEMPlate Entropy Regularisation (TEMPER), a post-hoc method that removes an attribute. TEMPER learns a linear map of the embedding in a split-latent variational autoencoder. The map is initialised at the principal component analysis (PCA) solution of the embeddings and trained to maximise the predictive entropy of an auxiliary attribute classifier. A relational term keeps the similarities of mated pairs close to the embedding similarities. We evaluate the removal of gender and a four-group ethnicity label from ArcFace embeddings against ten prior post-hoc methods and two label-free controls. Every method is trained on VGGFace2 and tested on IJB-C and RFW at false match rates down to 10−410^{-4}. Leakage is the balanced accuracy of the strongest of four attacker families trained on transformed VGGFace2 templates. TEMPER retains the highest IJB-C true match rate at a false match rate of 10−310^{-3} at every leakage threshold for both attributes. TEMPER retains a true match rate of 82.4% at a gender leakage of at most 52.5%, close to chance, against 76.7% for the best prior method. In the ablation, this improvement comes from the PCA initialisation and the relational term.

作者备注:Version 2: the method is renamed from VLEED to TEMPER and the title is changed. The method, the evaluation protocol and all results are revised compared to v1

AI 论文解读

基于摘要 · 非全文审阅

此版本尚未生成 AI 解读。解读由后台任务生成并按论文版本保存,登录后共享已有结果。

版本差异

本站首次收录为 v2,暂无更早版本可对照。

观测到的版本

相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。

v2 · 2026-11-17 02:21 · 最新观测版本

TEMPER: Reducing Attribute Leakage from Face Recognition Embeddings

Face recognition embeddings encode soft-biometric attributes such as gender and ethnicity, which can be inferred from templates. This paper proposes TEMPlate Entropy Regularisation (TEMPER), a post-hoc method that removes an attribute. TEMPER learns a linear map of the embedding in a split-latent variational autoencoder. The map is initialised at the principal component analysis (PCA) solution of the embeddings and trained to maximise the predictive entropy of an auxiliary attribute classifier. A relational term keeps the similarities of mated pairs close to the embedding similarities. We evaluate the removal of gender and a four-group ethnicity label from ArcFace embeddings against ten prior post-hoc methods and two label-free controls. Every method is trained on VGGFace2 and tested on IJB-C and RFW at false match rates down to 10−410^{-4}. Leakage is the balanced accuracy of the strongest of four attacker families trained on transformed VGGFace2 templates. TEMPER retains the highest IJB-C true match rate at a false match rate of 10−310^{-3} at every leakage threshold for both attributes. TEMPER retains a true match rate of 82.4% at a gender leakage of at most 52.5%, close to chance, against 76.7% for the best prior method. In the ablation, this improvement comes from the PCA initialisation and the relational term.

查看此版本解读 →查看此版本原文 ↗