Hand Gesture Recognition in Indian Sign Language Using Deep Learning

Abstract

Hand gestures serve as an essential means of communication for individuals with hearing and speech impairments. Indian Sign Language (ISL) is widely used by the deaf and hard-of-hearing community in India, yet a significant communication gap exists between ISL users and those unfamiliar with it. This research focuses on developing an efficient deep learning-based hand gesture recognition system for ISL to bridge this gap. The system utilizes Convolutional Neural Networks (CNNs) trained on a self-created dataset containing diverse ISL gestures. Unlike previous studies that primarily rely on controlled environments, our dataset incorporates real-world variations, making it more robust for practical applications. Furthermore, we introduce an innovative hybrid approach integrating CNNs with Recurrent Neural Networks (RNNs) to improve temporal gesture recognition, allowing for better classification of sequential signs. Experimental results demonstrate high accuracy and minimal loss, indicating the model’s potential for real-time deployment. This study also discusses challenges, limitations, and future improvements to enhance accuracy and adaptability in dynamic conditions.