Options
2025
Conference Paper
Title
Frequency Matters: Explaining Biases of Face Recognition in the Frequency Domain
Abstract
Face recognition (FR) models are vulnerable to performance variations across demographic groups. The causes for these performance differences are unclear due to the highly complex deep learning-based structure of face recognition models. Several works aimed at exploring possible roots of gender and ethnicity bias, identifying semantic reasons such as hairstyle, make-up, or facial hair as possible sources. Motivated by recent discoveries of the importance of frequency patterns in convolutional neural networks, we explain bias in face recognition using state-of-the-art frequency-based explanations. Our extensive results show that different frequencies are important to FR models depending on the ethnicity of the samples.
Conference
Keyword(s)
Branche: Information Technology
Research Line: Computer vision (CV)
Research Line: Human computer interaction (HCI)
Research Line: Machine learning (ML)
LTA: Interactive decision-making support and assistance systems
LTA: Machine intelligence, algorithms, and data structures (incl. semantics)
LTA: Generation, capture, processing, and output of images and 3D models
Biometrics
Face recognition
Machine learning
Computer vision
Deep learning
ATHENE