Unravelling the Effect of Image Distortions for Biased Prediction of Pre-trained Face Recognition Models

Identifying and mitigating bias in deep learning algorithms has gained significant popularity in the past few years due to its impact on the society. Researchers argue that models trained on balanced datasets with good representation provide equal and unbiased performance across subgroups. However,...

Full description

Saved in:
Bibliographic Details
Published in2021 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW) pp. 3779 - 3788
Main Authors Majumdar, Puspita, Mittal, Surbhi, Singh, Richa, Vatsa, Mayank
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.10.2021
Subjects
Analytical models
Computational modeling
Deep learning
Degradation
Face recognition
Nose
Systematics
Online AccessGet full text

Cover

More Information
Summary:Identifying and mitigating bias in deep learning algorithms has gained significant popularity in the past few years due to its impact on the society. Researchers argue that models trained on balanced datasets with good representation provide equal and unbiased performance across subgroups. However, can seemingly unbiased pre-trained model become biased when input data undergoes certain distortions? For the first time, we attempt to answer this question in the context of face recognition. We provide a systematic analysis to evaluate the performance of four state-of-the-art deep face recognition models in the presence of image distortions across different gender and race subgroups. We have observed that image distortions have a relationship with the performance gap of the model across different subgroups.
ISSN:2473-9944
DOI:10.1109/ICCVW54120.2021.00422