TU Darmstadt / ULB / TUbiblio

Deep Learning-based Face Recognition and the Robustness to Perspective Distortion

Damer, Naser and Wainakh, Yaza and Henniger, Olaf and Croll, Christian and Berthe, Benoit and Braun, Andreas and Kuijper, Arjan (2018):
Deep Learning-based Face Recognition and the Robustness to Perspective Distortion.
In: 2018 24th International Conference on Pattern Recognition (ICPR), Los Alamitos, IEEE, In: International Conference on Pattern Recognition (ICPR), Beijing, China, DOI: 10.1109/ICPR.2018.8545037,
[Online-Edition: https://doi.org/10.1109/ICPR.2018.8545037],
[Conference or Workshop Item]

Abstract

Face recognition technology is spreading into a wide range of applications. This is mainly driven by social acceptance and the performance boost achieved by the deep learningbased solutions in the recent years. Perspective distortion is an understudied distortion in face recognition that causes converging verticals when imaging 3D objects depending on the distance to the object. The effect of this distortion on face recognition was previously studied for algorithms based on hand-crafted features with a clear negative effect on verification performance. Possible solutions were proposed by compensating the distortion effect on the face image level, which requires knowing the camera settings and capturing a high quality image. This work investigates the effect of perspective distortion on the performance of a deep learning-based face recognition solution. It also provides a device parameter-independent solution to decrease this effect by creating more perspective-robust face representations. This was achieved by training the deep learning model on perspective-diverse data, without increasing the size of the training data. Experiments performed on the deep model in hand and a specifically collected database concluded that the perspective distortion effects face verification performance if not considered in the training process, and that this can be improved by our proposal of creating robust face representations by properly selecting the training data.

Item Type: Conference or Workshop Item
Erschienen: 2018
Creators: Damer, Naser and Wainakh, Yaza and Henniger, Olaf and Croll, Christian and Berthe, Benoit and Braun, Andreas and Kuijper, Arjan
Title: Deep Learning-based Face Recognition and the Robustness to Perspective Distortion
Language: English
Abstract:

Face recognition technology is spreading into a wide range of applications. This is mainly driven by social acceptance and the performance boost achieved by the deep learningbased solutions in the recent years. Perspective distortion is an understudied distortion in face recognition that causes converging verticals when imaging 3D objects depending on the distance to the object. The effect of this distortion on face recognition was previously studied for algorithms based on hand-crafted features with a clear negative effect on verification performance. Possible solutions were proposed by compensating the distortion effect on the face image level, which requires knowing the camera settings and capturing a high quality image. This work investigates the effect of perspective distortion on the performance of a deep learning-based face recognition solution. It also provides a device parameter-independent solution to decrease this effect by creating more perspective-robust face representations. This was achieved by training the deep learning model on perspective-diverse data, without increasing the size of the training data. Experiments performed on the deep model in hand and a specifically collected database concluded that the perspective distortion effects face verification performance if not considered in the training process, and that this can be improved by our proposal of creating robust face representations by properly selecting the training data.

Title of Book: 2018 24th International Conference on Pattern Recognition (ICPR)
Place of Publication: Los Alamitos
Publisher: IEEE
Uncontrolled Keywords: Face recognition, Biometrics, Biometric standards, Distortion
Divisions: 20 Department of Computer Science
20 Department of Computer Science > Mathematical and Applied Visual Computing
Event Title: International Conference on Pattern Recognition (ICPR)
Event Location: Beijing, China
Date Deposited: 26 Jun 2019 07:45
DOI: 10.1109/ICPR.2018.8545037
Official URL: https://doi.org/10.1109/ICPR.2018.8545037
Export:

Optionen (nur für Redakteure)

View Item View Item