Kinect-based automatic sign language recognition

• Main Author: Lin, Zi Ying
• Other Authors: Cham Tat Jen
• Format: Final Year Project
• Language: English
• Published: 2016
• Subjects: DRNTU::Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision
• Online Access: http://hdl.handle.net/10356/66915
• Institution: Nanyang Technological University

As technology advances, so does its popularity as a teaching tool, and it can be envisioned that technology will play an integral role in education in the future. This project aims to integrate the use of technology with sign-language to help improve and promote communication not only amongst the hearing-impaired, but also with their friends and family as well. The project works on using the functionality of the depth sensor on the Microsoft Kinect to recognize gestures according to the American Sign Language (ASL). The project is a prototype that demonstrates the concept of learning with technology by allowing users to practice the ASL hand gestures through an educational game. The depth sensor in the Microsoft Kinect allows the project to detect and identify the hand and 5 individual fingers. The concept of gesture recognition is essential to future developments of this project and serves a basis, as this project scratches only the surface of the potential of this concept. This report details the process of development of this project, the detection of the hand and fingers, as well as the recognition of gestures.