Dan Green

Research project title: 
Representing Signed Language for Automated Understanding
Principal goal for project: 
To robustly represent signed language in a machine readable form in order to examine signed language within NLP.
Research project: 

Sign languages have no written form and are typically fed into machines as 2D videos which are extremely difficult to analyse directly. By creating a robust representation schema for signed language it will be possible to better define the computer vision problem of working with videos and allow well-founded NLP projects to work with signed language.

About me: 

At undergraduate level I have degrees in both Computer Science (University of Cambridge) and Sign Language Interpreting (Heriot-Watt). I combined these both in my Data Science MSc (Heriot-Watt) where my research project examined automated sign language recognition.

Supervisor: 
Student type: 
Current student