Monday 07 Mar 2016: Automated Sign Language Recognition using Discriminative Sequential Pattern Trees
Nicolas Pugeault -
Automated Sign Language Recognition (SLR) remains a challenging problem to this day. Like spoken languages, sign language feature thousands of signs, sometimes only differing by subtle changes in hand motion, shape or position. This, compounded with differences in signing style and physiology between individuals, makes SLR an intricate challenge. A common approach for learning temporal sequences that form a sign is to use Hidden Markov Models, but they are very inefficient both at training and detection time when applied on large dictionaries. Additionally, generalising over signer specific variations require enormous amount of data. In this talk I will introduce an approach inspired from the data mining literature based on “Sequential Patterns”, and present novel algorithms for learning efficiently discriminative patterns to detect and recognise individual signs.