Real Time Sign Language Recognition Using Computer Vision And Ai

This project presents a real-time American Sign Language (ASL) recognition system leveraging the YOLOv8 object detection framework integrated with a Streamlit-based web interface.Designed to enhance accessibility for the deaf and hard-of-hearing community, the system processes live webcam input using OpenCV, performs ASL gesture detection via a lightweight YOLOv8n model trained on a custom-labeled dataset, and displays annotated results in real time.The technology stack includes Python, PyTorch, OpenCV, and the Ultralytics YOLOv8 API for model training and inference, while the user interface is built with Streamlit for seamless interaction.Achieving high accuracy ([email protected] 94%) and near real-time performance (~28 FPS), the system demonstrates strong potential for deployment in educational, communicative, and assistive applications.Its modular design supports extensibility to broader sign language vocabularies and dynamic gesture sequences.This project supports SDG Goal 10: Reduced Inequalities, by promoting accessible communication tools for the deaf and hard-of-hearing community.By leveraging technology to bridge language gaps, it empowers individuals with disabilities to participate more fully in education, employment, and social interaction.The system also aligns with SDG Goal 4: Quality Education, as it can be used as a learning tool for both sign language users and learners.Additionally, it promotes SDG Goal 9: Industry, Innovation and Infrastructure by utilizing advanced AI and computer vision technologies for social impact.Furthermore, our project aligns with academic goals and successfully satisfies Program Outcomes (PO) Level 1 to Level 8 and Program Specific Outcomes (PSO) Level 1 and Level 2, ensuring both technical competence and societal relevance.SDG 3: Good Health and Well-being