/RealTimeSignLanguageRecognition

Applied SSD integrated with MobileNet model for object (sign gestures) detection and recognition and the model is trained using Transfer Learning, with the aim to develop a web app for real-time ASL recognition from user input & then to generate text in English.

Primary LanguageJupyter Notebook

Real Time Sign Language Recognition using SSD and MobileNetv2

This project applies Single-Shot Multibox Detector (SSD) architecture integrated with the mobilenetv2 model for the object(sign gestures) detection and recognition of Americal Sign Langguage (ASL) using Transfer Learning, with the aim to develop a web app for real-time ASL recognition from user input through vide frame & then to generate text in English.

The web App for neural network model is deployed in React and Flask.