Continuous Sign Language Translation Using Deep Learning

This research focuses on developing an end-to-end deep learning framework for Continuous Sign Language Translation (CSL T), addressing the challenge of converting sign language videos into coherent textual output. Our approach integrates spatial and temporal feature extraction techniques to effectively capture hand movements and linguistic structures. Initially, a Convolutional Neural Network (CNN) extracts spatial features from video frames, while a Recurrent Neural Network (RNN) models temporal dependencies across consecutive frames. To enhance translation fluency, we incorporate a Transformer-based Neural Machine Translation (NMT) model, improving the conversion of sign glosses into natural language.