Deep eigen-filters for face recognition: Feature representation via unsupervised multi-structure filter learning. (April 2020)

• Record Type: Journal Article
• Authors: Zhang, Ming; Khan, Sheheryar; Yan, Hong
• Is Part Of: Pattern recognition. Volume 100(2020:Apr.)
• Publication Date: 2020-04
• Subjects: Deep eigen-filters -- Convolution kernels -- Face recognition -- Convolutional neural networks -- Feature representation; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2019.107176
• Languages: English
• ISSNs: 0031-3203

Abstract

Highlights: Propose a three-stage multi-structure filter learning approach inspired from advances in convolutional layers of convolutional neural networks. Analyze the linear combination between obtained filters and convolution kernels in convolutional neural networks for filter selection. Build a network for feature representation based on learned filters. Competitive face recognition performance with less computational cost and high robustness to facial expression and illumination compared to other deep learning-based methods. Abstract: Training deep convolutional neural networks (CNNs) often requires high computational cost and a large number of learnable parameters. To overcome this limitation, one solution is computing predefined convolution kernels from training data. In this paper, we propose a novel three-stage approach for filter learning alternatively. It learns filters in multiple structures including standard filters, channel-wise filters and point-wise filters which are inspired from variations of CNNs' convolution operations. By analyzing the linear combination between learned filters and original convolution kernels in pre-trained CNNs, the reconstruction error is minimized to determine the most representative filters from the filter bank. These filters are used to build a network followed by HOG-based feature extraction for feature representation. The proposed approach shows competitive performance on color face recognition compared with other deep CNNs-based methods. Besides, it provides a perspective of interpreting CNNs by introducing the concepts of advanced convolutional layers to unsupervised filter learning.