Robust Sign Language Recognition System Using ToF Depth Cameras

Abstract: Sign language recognition is a difficult task, yet required for many applications in real-time speed. Using RGB cameras for recognition of sign languages is not very successful in practical situations and accurate 3D imaging requires expensive and complex instruments. With introduction of Time-of-Flight (ToF) depth cameras in recent years, it has become easier to scan the environment for accurate, yet fast depth images of the objects without the need of any extra calibrating object. In this paper, a robust system for sign language recognition using ToF depth cameras is presented for converting the recorded signs to a standard and portable XML sign language named SiGML for easy transferring and converting to real-time 3D virtual characters animations. Feature extraction using moments and classification using nearest neighbor classifier are used to track hand gestures and significant result of 100% is achieved for the proposed approach.