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{196365,
        author = {Y.P. Hari Krishna and M. Venkata Pranay and S. kavya and P. Sindhu Vaishnavi and P. Bhavitha Devi Sri},
        title = {Sign Language Translation: Computer Vision and Deep Learning Approach for Gesture Recognition},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {11},
        pages = {3227-3231},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=196365},
        abstract = {The present project is devoted to creation of intelligent, real-time sign language translator based on computer vision and deep learning methods of gesture recognition. Sign language is an important communication tool among the hearing and speech impaired people. But the majority lacks the knowledge on what it entails and this poses a great difference in communication. The purpose of this system is to bridge that gap by automatically identifying the hand gestures and translating them into useful text or speech. The suggested system is a vision-based system, and the wearable devices such as sensor-based gloves are eliminated. Webcam captures real-time video input that passes through a number of processes, such as preprocessing, hand detection, feature extraction, and classification. State-of-the-art deep learning algorithms, such as Convolutional Neural Networks (CNN) are used to synthesize spatial gestures. In the meantime, Long Short-Term Memory (LSTM) networks are used to capture the patterns in dynamic gestures. The data comprises both the stationary and the moving gestures recorded in different settings. Normalization, background subtraction and noise reduction methods allow preprocessing to enhance the quality of data. The trained model is able to categorize gestures correctly and translate them into text which can be converted to speech using text to speech systems. The system is accurate, real time and resistant to various lighting and background conditions. The project provides a solution to assistive communication which is scalable and cost effective and has possible uses in education, healthcare and human computer interaction systems.},
        keywords = {Sign Language, Gesture Recognition, Computer Vision, Deep Learning, CNN, LSTM, Human-Computer Interaction.},
        month = {April},
        }