AnAgent-Based Multimodal Interface forSketchInterpretation

We present amultimodal interface forsketch inter- pretation thatrelies onamulti-agent architecture. Thedesign oftheinterpretation engine andthedifferent agents arebased onauser-centered approach whereefficiency measureisdefined asusersatisfaction. Sofar, several graphical agents havebeen implemented forrecognizing basic graphical objects (e.g., lines, circles, etc) aswell asmorecomplex (e.g., hatches, stairs, captions, etc)inarchitectural design. Besides, vocalagentshavebeen developed forrecognizing spokenannotations (e.g., dimensions) andinterface commands. Realistic evaluations withprofessional usershavedemonstrated thepotential interest oftheproposed system.

[1]  Sharon L. Oviatt,et al.  Multimodal Integration - A Statistical View , 1999, IEEE Trans. Multim..

[2]  Pierre Leclercq,et al.  A freehand-sketch environment for architectural design supported by a multi-agent system , 2005, Comput. Graph..

[3]  Erkki Oja,et al.  On-line adaptation in recognition of handwritten alphanumeric characters , 1999, Proceedings of the Fifth International Conference on Document Analysis and Recognition. ICDAR '99 (Cat. No.PR00318).