Automatic sign language detector for video call

• Main Author: Chua, Mark De Wen
• Other Authors: Lam Siew Kei
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2021
• Subjects: Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision; Engineering::Computer science and engineering::Computing methodologies::Pattern recognition
• Online Access: https://hdl.handle.net/10356/148038
• Institution: Nanyang Technological University

Video conference has been a big part of our lives since COVID-19 hit but the hearing-impaired does not have the ability to communicate in an efficient way when video conferencing. Singapore Association For The Deaf (SADeaf) state that there was a rise of interest to learn sign language for communication with hearing-impaired family member or co-workers. However, there is a steep learning curve for learning sign language. This project aims to allow real-time interpretation of sign language using You-Only-Look-Once(YOLO) neural networks. The application will be designed to output the word visually and audibly when the user uses sign language on their web camera while video conferencing.