SIGN LANGUAGE RECOGNITION

  • Motivation: This work develops sign language recognition through computer vision techniques to bridge communication barriers faced by the deaf and hard-of-hearing community.
  • Objective: • To recognize American sign language and get a text and audio output.
  • Project date: 17 December, 2023
  • Project URL: GitHub Repository

Abstract

To accomplish sign language recognition, initially crucial step is to detect key points using the MP holistic library. Key-point video data is collected using an open CV. This 50-video data for each gesture is collected, with 30 frames each. Using this data, a deep neural network model is trained to provide visual output confidence and auditory output with 90% accuracy.