High Dimensional Computing Approach to Detection and Learning Gesture Biometrics

Gesture data has emerged as a pivotal element in shaping the future of technology, influencing human-computer interaction. In this research paper, we explore biometric gesture recognition, recognizing the inherent significance of gesture data as a biometric identifier with potential applications in detection of human motions and intentions. The primary focus of our work revolves around the deployment of the Hyperdimensional Computing (HDC) methodology to create a machine-learning model that is tailored for gesture recognition. Our study showcases the efficacy of HDC in detecting and classifying gestures across two critical problem areas: gesture identification and biometric identification. Using the Microsoft Kinect Gesture Data Set, our experiments demonstrate the robustness of HDC in capturing a variety of gesture patterns. The dual application of HDC in both recognizing specific gestures and establishing biometric identity underscores its versatility and potential in diverse scenarios. The results presented in this paper highlight the promising approach of HDC in the context of gesture recognition. As we conclude, we discuss the implications of our findings and outline directions for future work.