American Sign Language Recognition and Conversion using CNN

American Sign Language Recognition and Conversion using CNN

Aditya Thakar1, Ameya Katgeri2, Devdatta Patil3,

Warepam Richard Singh4, Prof. P. S. Jadhavar5

1,2,3,4 Student, Department Of Computer Engineering, P.V.P.I.T, Pune, Maharashtra, India

5 Professor, Department Of Computer Engineering, P.V.P.I.T, Pune, Maharashtra, India.

---------------------------------------------------------------------***---------------------------------------------------------------------

Abstract - This paper discusses the development of an application that can recognize and translate American Sign Language (ASL) gestures into text to facilitate communication between ASL users and non-native speakers. The application utilizes computer vision techniques and natural language processing, including convolutional neural networks and data augmentation techniques, to interpret and translate ASL gestures into text. The paper explores the challenges in ASL recognition due to variability in signing styles, dialects, and lighting conditions. The data for training the model was collected manually using OpenCV and the HandTrackingModule. The paper highlights the advantages and limitations of existing approaches and provides a comparative study of various works in recognizing ASL using vision-based techniques.

Key Words:  ASL translation, Deep Learning, CNN, Mediapipe