Egocentric Gesture Recognition for Head-Mounted AR devices

TL;DR

This paper introduces a deep learning framework for egocentric gesture recognition in AR devices, utilizing a novel data augmentation method and a new dataset, achieving improved accuracy over existing methods.

Contribution

It presents an end-to-end deep learning approach combining ego-hand encoding and recurrent networks, along with a novel green screen data augmentation technique and a new gesture dataset.

Findings

Achieved higher recognition accuracy compared to state-of-the-art methods.

Developed a novel data augmentation technique using green screen capture.

Published a new dataset of egocentric gestures in controlled and natural environments.

Abstract

Natural interaction with virtual objects in AR/VR environments makes for a smooth user experience. Gestures are a natural extension from real world to augmented space to achieve these interactions. Finding discriminating spatio-temporal features relevant to gestures and hands in ego-view is the primary challenge for recognising egocentric gestures. In this work we propose a data driven end-to-end deep learning approach to address the problem of egocentric gesture recognition, which combines an ego-hand encoder network to find ego-hand features, and a recurrent neural network to discern temporally discriminating features. Since deep learning networks are data intensive, we propose a novel data augmentation technique using green screen capture to alleviate the problem of ground truth annotation. In addition we publish a dataset of 10 gestures performed in a natural fashion in front of a green screen for training and the same 10 gestures performed in different natural scenes without green screen for validation. We also present the results of our network's performance in comparison to the state-of-the-art using the AirGest dataset