Automated hand gesture recognition for enhancing sign language communication

This paper introduces a novel approach aimed at enhancing communication between individuals who are deaf or hard of hearing and those unfamiliar with sign language. The project addresses this challenge by developing a mobile application that harnesses the power of smartphone cameras, coupled with a...

Full description

Bibliographic Details
Main Author: Lee, Teck Junn
Format: Final Year Project / Dissertation / Thesis
Published: 2024
Subjects:
Online Access:http://eprints.utar.edu.my/6654/
http://eprints.utar.edu.my/6654/1/fyp_CS_2024_LTJ.pdf
Description
Summary:This paper introduces a novel approach aimed at enhancing communication between individuals who are deaf or hard of hearing and those unfamiliar with sign language. The project addresses this challenge by developing a mobile application that harnesses the power of smartphone cameras, coupled with a deep learning model, to interpret hand gestures and provide real-time contextual information to users. It emphasizes the widespread adoption of smartphones and the practical applicability of mobile applications in real-life scenarios. Furthermore, the paper proposes a new methodology leveraging Google’s MediaPipe, which outperforms traditional approaches such as transfer learning with pre-trained object detection models in deep learning model development. Of paramount importance is the seamless integration of the deep learning model with the mobile application, enabling real-time detection and recognition on the mobile application.