Sign Language Fingerspelling Classification from Depth and Color Images using a Deep Belief Network

TL;DR

This paper introduces a new feature extraction method for American Sign Language fingerspelling recognition using depth and color images from a Kinect, achieving high accuracy and real-time classification.

Contribution

The paper presents a novel feature extraction technique tailored for hand pose recognition with depth and intensity images, applied to sign language classification using a Deep Belief Network.

Findings

Achieved 99% precision and recall with known users.

Achieved 77% recall and 79% precision with unseen users.

Method supports real-time sign classification and environmental adaptability.

Abstract

Automatic sign language recognition is an open problem that has received a lot of attention recently, not only because of its usefulness to signers, but also due to the numerous applications a sign classifier can have. In this article, we present a new feature extraction technique for hand pose recognition using depth and intensity images captured from a Microsoft Kinect sensor. We applied our technique to American Sign Language fingerspelling classification using a Deep Belief Network, for which our feature extraction technique is tailored. We evaluated our results on a multi-user data set with two scenarios: one with all known users and one with an unseen user. We achieved 99% recall and precision on the first, and 77% recall and 79% precision on the second. Our method is also capable of real-time sign classification and is adaptive to any environment or lightning intensity.