Force myography benchmark data for hand gesture recognition and transfer learning

TL;DR

This paper introduces a publicly available force myography dataset for hand gesture recognition, enabling standardized benchmarking and demonstrating how transfer learning can improve recognition accuracy across different users.

Contribution

The authors provide a new benchmark dataset for force myography-based hand gesture recognition, facilitating algorithm comparison and research on transfer learning methods.

Findings

Transfer learning improves gesture recognition accuracy across users.

The dataset includes 18 gestures from 20 individuals, enabling diverse research.

Benchmarking results demonstrate the dataset's utility for transfer learning studies.

Abstract

Force myography has recently gained increasing attention for hand gesture recognition tasks. However, there is a lack of publicly available benchmark data, with most existing studies collecting their own data often with custom hardware and for varying sets of gestures. This limits the ability to compare various algorithms, as well as the possibility for research to be done without first needing to collect data oneself. We contribute to the advancement of this field by making accessible a benchmark dataset collected using a commercially available sensor setup from 20 persons covering 18 unique gestures, in the hope of allowing further comparison of results as well as easier entry into this field of research. We illustrate one use-case for such data, showing how we can improve gesture recognition accuracy by utilising transfer learning to incorporate data from multiple other persons. This also illustrates that the dataset can serve as a benchmark dataset to facilitate research on transfer learning algorithms.