Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{202139,
        author = {Harshitha G.J and Dr Manjunatha B H},
        title = {Kannada sign language Recognition using deep learning},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {12},
        pages = {6979-6990},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=202139},
        abstract = {Sign language recognition has emerged as a crucial research area in assistive computing, aiming to bridge the communication gap between hearing-impaired individuals and the general population. For individuals with hearing and speech disabilities, sign language serves as the primary means of communication, supporting effective interaction in daily life. However, the absence of widespread sign language knowledge among the general population often leads to communication difficulties, emphasizing the need for automated and intelligent translation systems. Although significant research has been conducted on American Sign Language (ASL) and Indian Sign Language (ISL), regional sign languages such as Kannada Sign Language (KSL) remain largely unexplored. Recent advancements in computer vision and deep learning have enabled the development of camera-based sign language recognition systems that are non-intrusive, cost-effective, and suitable for real-time deployment. This under representation is mainly due to the lack of standardized datasets, limited linguistic resources, and insufficient region-specific recognition frameworks. Consequently, existing models often fail to generalize well to KSL, highlighting the necessity for focused research in this area. Image-based Convolutional Neural Networks (CNNs) are widely used for extracting spatial features from hand gestures, while landmark-based approaches using MediaPipe provide accurate hand keypoint detection with reduced computational complexity. Additionally, sequential models such as Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) networks effectively capture temporal variations in dynamic gestures, and transformer-based methods offer improved modeling of complex gesture patterns. This literature survey reviews and compares these approaches, identifies key challenges, and highlights research gaps, thereby motivating the development of a real-time, deep learning-driven Kannada Sign Language recognition system. Such a system can enhance accessibility, promote inclusive communication, and support practical applications in education, public services, and human– computer interaction. By focusing on accuracy, efficiency, and real-time performance, future research can significantly improve the quality of life for the hearing-impaired community.},
        keywords = {Kannada Sign Language, Deep Learning, MediaPipe, CNN, LSTM, Gesture Recognition, Assistive Systems.},
        month = {May},
        }