Paper: Interpreting Pointing Gestures and Spoken Requests – A Probabilistic Salience-based Approach

ACL ID C10-2178
Title Interpreting Pointing Gestures and Spoken Requests – A Probabilistic Salience-based Approach
Venue International Conference on Computational Linguistics
Session Poster Session
Year 2010
Authors

We present a probabilistic, salience-based approach to the interpretation of pointing gestures together with spoken utterances. Our mechanism models dependencies be- tween spatial and temporal aspects of ges- tures and features of utterances. For our evaluation, we collected a corpus of re- quests which optionally included point- ing. Our results show that pointing infor- mation improves interpretation accuracy.