Deep Fisher Discriminant Learning for Mobile Hand Gesture Recognition

TL;DR

This paper introduces Fisher discriminative deep models, F-BLSTM and F-BGRU, which integrate Fisher criterion into recurrent neural networks for improved mobile gesture recognition, validated on a large new database and benchmark datasets.

Contribution

It proposes novel deep models F-BLSTM and F-BGRU that incorporate Fisher criterion for enhanced gesture classification accuracy.

Findings

Fisher discriminative models outperform traditional BLSTM and BGRU.

Models achieve superior accuracy on MGD, BUAA, and SmartWatch datasets.

Extensive experiments confirm the effectiveness of the proposed approach.

Abstract

Gesture recognition is a challenging problem in the field of biometrics. In this paper, we integrate Fisher criterion into Bidirectional Long-Short Term Memory (BLSTM) network and Bidirectional Gated Recurrent Unit (BGRU),thus leading to two new deep models termed as F-BLSTM and F-BGRU. BothFisher discriminative deep models can effectively classify the gesture based on analyzing the acceleration and angular velocity data of the human gestures. Moreover, we collect a large Mobile Gesture Database (MGD) based on the accelerations and angular velocities containing 5547 sequences of 12 gestures. Extensive experiments are conducted to validate the superior performance of the proposed networks as compared to the state-of-the-art BLSTM and BGRU on MGD database and two benchmark databases (i.e. BUAA mobile gesture and SmartWatch gesture).