Bangla Hand Sign Recognition Using HANDCNN Model for Multi-Hand Sign Datasets

Abstract

Hand sign recognition is an emerging field in computer vision that focuses on understanding and interpreting gestures made by human hands. This technology plays a crucial role in various applications, including sign language interpretation, human-computer interaction, and virtual reality. In this study, we propose a novel approach for hand sign recognition using deep learning techniques. In our method of HANDCNN model spatial and temporal information are present in hand gestures. By leveraging a large dataset of hand sign images, we train our model to accurately recognize and classify a wide range of hand signs. We conduct extensive experiments to evaluate the performance of our approach and compare it with existing methods. The results demonstrate the effectiveness and robustness of our proposed approach, achieving state-of-the-art performance in hand sign recognition. Furthermore, we discuss potential applications and future directions for improving hand sign recognition systems, highlighting the importance of this technology in facilitating inclusive communication and enhancing human-machine interaction.