Dynamic Hand Gesture Recognition for Wearable Devices with Low Complexity Recurrent Neural Networks

TL;DR

This paper introduces two low-complexity recurrent neural network-based methods for dynamic hand gesture recognition in wearable devices, utilizing CNN-RNN and accelerometer data with fixed-point optimization for efficiency.

Contribution

It presents novel low-complexity RNN algorithms for gesture recognition, combining CNNs with RNNs and optimizing for hardware efficiency in wearable devices.

Findings

CNN-RNN method achieves high accuracy on video signals.

Accelerometer-based RNN method is lightweight and efficient.

Fixed-point optimization reduces memory and power consumption.

Abstract

Gesture recognition is a very essential technology for many wearable devices. While previous algorithms are mostly based on statistical methods including the hidden Markov model, we develop two dynamic hand gesture recognition techniques using low complexity recurrent neural network (RNN) algorithms. One is based on video signal and employs a combined structure of a convolutional neural network (CNN) and an RNN. The other uses accelerometer data and only requires an RNN. Fixed-point optimization that quantizes most of the weights into two bits is conducted to optimize the amount of memory size for weight storage and reduce the power consumption in hardware and software based implementations.