Designed a Interpreter to convert Sign-Language video into text to reduce Communication barrier between mute, hard of hearing, and normal person.
- Leverages a Transformer Architecture with DenseNet121-based feature extractor, enhanced by an accessible app.
- Compiled and curated an extensive dataset of 1600+ Indian sign language videos, enabling accurate recognition and classification of up to 4 distinct signing gestures, revolutionizing accessibility technology.
Watch a demo of the Sign-Language Interpreter in action: