Towards Sign language recognition system in Human-Computer interactions

摘要

This paper presents a modeling approach for Sign language recognition which is introduced through a dialogue between deaf people and signing avatar. By this modeling approach, the system can be configurable as the scenario and the vocabulary can be modified or changed. Tow important elements are included to these modeling: the context and the prediction. The objective is to improve the reliability of sign language recognition system compared to the classic systems which generally do not use semantic concept.