Incorporating Facial Features into a Multi-Channel Gesture Recognition System for the Interpretation of Irish Sign Language Sequences

摘要

In this paper we present a novel gesture recognition system for the interpretation of Irish Sign Language sequences which incorporates manual and non-manual information. We implement a set of independent Hidden Markov Model networks to recognize hand gestures, head movements and facial features into a single framework for interpreting Irish Sign Language. This framework is not specific to any particular type of gesture and we demonstrate this by showing that manual and non manual signals can be robustly spotted and classified from with continuous sign sequences.