Our deep cnn face matchers have developed achromatopsia

A Bhatta, D Mery, H Wu, J Annan… - Proceedings of the …, 2024 - openaccess.thecvf.com
Modern deep CNN face matchers are trained on datasets containing" color" images. We
show that such matchers achieve essentially the same accuracy on color images when …

Facial hair area in face recognition across demographics: Small size, big effect

H Wu, S Tian, A Bhatta, K Öztürk… - Proceedings of the …, 2024 - openaccess.thecvf.com
Observed variations in face recognition accuracy across demographics, often viewed as"
bias", have motivated research into the causes of such variations. Variations in facial …

Can the accuracy bias by facial hairstyle be reduced through balancing the training data?

K Ozturk, H Wu, KW Bowyer - Proceedings of the IEEE/CVF …, 2024 - openaccess.thecvf.com
Appearance of a face can be greatly altered by growing a beard and mustache. The facial
hairstyles in a pair of images can cause marked changes to the impostor distribution and the …

LogicNet: A Logical Consistency Embedded Face Attribute Learning Network

H Wu, S Tian, H Li, KW Bowyer - arXiv preprint arXiv:2311.11208, 2023 - arxiv.org
Ensuring logical consistency in predictions is a crucial yet overlooked aspect in multi-
attribute classification. We explore the potential reasons for this oversight and introduce two …