Compact CNN Models for On-device Ocular-based User Recognition in Mobile Devices

TL;DR

This paper evaluates neural network pruning and knowledge distillation techniques to develop compact CNN models for ocular-based user recognition on resource-limited mobile devices, demonstrating improved accuracy and efficiency.

Contribution

It is the first to compare five pruning methods with knowledge distillation for mobile ocular recognition, highlighting the superiority of knowledge distillation in accuracy and inference speed.

Findings

Layerwise magnitude-based pruning achieves 8x compression with ResNet50.

Knowledge distillation outperforms pruning in verification accuracy.

Distilled models enable real-time inference on various mobile devices.

Abstract

A number of studies have demonstrated the efficacy of deep learning convolutional neural network (CNN) models for ocular-based user recognition in mobile devices. However, these high-performing networks have enormous space and computational complexity due to the millions of parameters and computations involved. These requirements make the deployment of deep learning models to resource-constrained mobile devices challenging. To this end, only a handful of studies based on knowledge distillation and patch-based models have been proposed to obtain compact size CNN models for ocular recognition in the mobile environment. In order to further advance the state-of-the-art, this study for the first time evaluates five neural network pruning methods and compares them with the knowledge distillation method for on-device CNN inference and mobile user verification using ocular images. Subject-independent analysis on VISOB and UPFR-Periocular datasets suggest the efficacy of layerwise magnitude-based pruning at a compression rate of 8 for mobile ocular-based authentication using ResNet50 as the base model. Further, comparison with the knowledge distillation suggests the efficacy of knowledge distillation over pruning methods in terms of verification accuracy and the real-time inference measured as deep feature extraction time on five mobile devices, namely, iPhone 6, iPhone X, iPhone XR, iPad Air 2 and iPad 7th Generation.