Comparing Robustness of Pairwise and Multiclass Neural-Network Systems for Face Recognition

• Published in: EURASIP journal on advances in signal processing Vol. 2008; no. 1
• Main Authors: Uglov, J., Jakaite, L., Schetinin, V., Maple, C.
• Format: Journal Article
• Language: English
• Published: New York Springer Nature B.V 01-01-2008; SpringerOpen
• Subjects: Biometrics; Experiments; Neural networks; Noise; Standard deviation; Studies
• ISSN: 1687-6180; 1687-6172; 1687-6180
• DOI: 10.1155/2008/468693

Noise, corruptions, and variations in face images can seriously hurt the performance of face-recognition systems. To make these systems robust to noise and corruptions in image data, multiclass neural networks capable of learning from noisy data have been suggested. However on large face datasets such systems cannot provide the robustness at a high level. In this paper, we explore a pairwise neural-network system as an alternative approach to improve the robustness of face recognition. In our experiments, the pairwise recognition system is shown to outperform the multiclass-recognition system in terms of the predictive accuracy on the test face images.