Distill and De-bias: Mitigating Bias in Face Verification using Knowledge Distillation
원문 초록
Face recognition networks generally demonstrate bias with respect to sensitive attributes like gender, skintone etc. For gender and skintone, we observe that the regions of the face that a network attends to vary by the category of an attribute. This might contribute to bias. Building on this intuition, we propose a novel distillation-based approach called Distill and De-bias (D&D) to enforce a network to attend to similar face regions, irrespective of the attribute category. In D&D, we train a teacher network on images from one category of an attribute; e.g. light skintone. Then distilling information from the teacher, we train a student network on images of the remaining category; e.g., dark skintone. A feature-level distillation loss constrains the student network to generate teacher-like representations. This allows the student network to attend to similar face regions for all attribute categories and enables it to reduce bias. We also propose a second distillation step on top of D&D, called D&D++. Here, we distill the `un-biasedness' of the D&D network into a new student network, the D&D++ network, while training this new network on all attribute categories; e.g., both light and dark skintones. This helps us train a network that is less biased for an attribute, while obtaining higher face verification performance than D&D. We show that D&D++ outperforms existing baselines in reducing gender and skintone bias on the IJB-C dataset, while obtaining higher face verification performance than existing adversarial de-biasing methods. We evaluate the effectiveness of our proposed methods on two state-of-the-art face recognition networks: ArcFace and Crystalface.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
Rectifying the Data Bias in Knowledge Distillation
Attributes Shape the Embedding Space of Face Recognition Models
The Trilemma of Truth in Large Language Models
Non-Adaptive Adversarial Face Generation
Benchmarking Foundation Models for Zero-Shot Biometric Tasks
50 Years of Automated Face Recognition
토론
이 논문과 연결된 포럼 글이 아직 없습니다.