When Face Recognition Meets with Deep Learning: an Evaluation of Convolutional Neural Networks for Face Recognition

TL;DR

This paper evaluates CNN-based face recognition systems trained on the public LFW dataset, proposing new architectures, analyzing implementation choices, and highlighting the importance of CNN fusion and metric learning for improved performance.

Contribution

It introduces three CNN architectures trained on LFW, compares different design choices, and provides insights into effective properties like feature dimensionality reduction and fusion strategies.

Findings

Reducing feature dimensionality does not harm accuracy.

Fusion of multiple CNNs enhances face recognition performance.

Metric learning significantly improves CNN-based face recognition.

Abstract

Deep learning, in particular Convolutional Neural Network (CNN), has achieved promising results in face recognition recently. However, it remains an open question: why CNNs work well and how to design a 'good' architecture. The existing works tend to focus on reporting CNN architectures that work well for face recognition rather than investigate the reason. In this work, we conduct an extensive evaluation of CNN-based face recognition systems (CNN-FRS) on a common ground to make our work easily reproducible. Specifically, we use public database LFW (Labeled Faces in the Wild) to train CNNs, unlike most existing CNNs trained on private databases. We propose three CNN architectures which are the first reported architectures trained using LFW data. This paper quantitatively compares the architectures of CNNs and evaluate the effect of different implementation choices. We identify several useful properties of CNN-FRS. For instance, the dimensionality of the learned features can be significantly reduced without adverse effect on face recognition accuracy. In addition, traditional metric learning method exploiting CNN-learned features is evaluated. Experiments show two crucial factors to good CNN-FRS performance are the fusion of multiple CNNs and metric learning. To make our work reproducible, source code and models will be made publicly available.