Binarized Neural Networks for Resource-Constrained On-Device Gait Identification

TL;DR

This paper introduces BiPedalNet, a binarized neural network that enables efficient on-device gait recognition with significantly reduced memory requirements while maintaining high accuracy.

Contribution

The paper presents a novel binarized CNN architecture, BiPedalNet, optimized for resource-constrained mobile devices for gait-based user authentication.

Findings

BiPedalNet nearly matches state-of-the-art accuracy on the Padova gait dataset.

It reduces memory overhead to 1/32 of traditional models.

Demonstrates feasibility of on-device gait recognition with low-resource hardware.

Abstract

User authentication through gait analysis is a promising application of discriminative neural networks -- particularly due to the ubiquity of the primary sources of gait accelerometry, in-pocket cellphones. However, conventional machine learning models are often too large and computationally expensive to enable inference on low-resource mobile devices. We propose that binarized neural networks can act as robust discriminators, maintaining both an acceptable level of accuracy while also dramatically decreasing memory requirements, thereby enabling on-device inference. To this end, we propose BiPedalNet, a compact CNN that nearly matches the state-of-the-art on the Padova gait dataset, with only 1/32 of the memory overhead.