19 June 2017 Finessing filter scarcity problem in face recognition via multi-fold filter convolution
Author Affiliations +
Proceedings Volume 10443, Second International Workshop on Pattern Recognition; 104430G (2017) https://doi.org/10.1117/12.2280352
Event: Second International Workshop on Pattern Recognition, 2017, Singapore, Singapore
Abstract
The deep convolutional neural networks for face recognition, from DeepFace to the recent FaceNet, demand a sufficiently large volume of filters for feature extraction, in addition to being deep. The shallow filter-bank approaches, e.g., principal component analysis network (PCANet), binarized statistical image features (BSIF), and other analogous variants, endure the filter scarcity problem that not all PCA and ICA filters available are discriminative to abstract noise-free features. This paper extends our previous work on multi-fold filter convolution (-FFC), where the pre-learned PCA and ICA filter sets are exponentially diversified by folds to instantiate PCA, ICA, and PCA-ICA offspring. The experimental results unveil that the 2-FFC operation solves the filter scarcity state. The 2-FFC descriptors are also evidenced to be superior to that of PCANet, BSIF, and other face descriptors, in terms of rank-1 identification rate (%).
© (2017) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Cheng-Yaw Low, Andrew Beng-Jin Teoh, "Finessing filter scarcity problem in face recognition via multi-fold filter convolution", Proc. SPIE 10443, Second International Workshop on Pattern Recognition, 104430G (19 June 2017); doi: 10.1117/12.2280352; https://doi.org/10.1117/12.2280352
PROCEEDINGS
5 PAGES


SHARE
Back to Top