Vincent Delfosse

An Agent-Based Multimodal Interface for Sketch Interpretation

We present a multimodal interface for sketch interpretation that relies on a multi-agent architecture. The design of the interpretation engine and the different agents are based on a user-centered approach where efficiency measure is defined as user satisfaction. So far, several graphical agents have been implemented for recognizing basic graphical objects (e.g. lines, circles, etc.) as well as more complex (e.g., hatches, stairs, captions, etc) in architectural design. Besides, vocal agents have been developed for recognizing spoken annotations (e.g. dimensions) and interface commands. Realistic evaluations with professional users have demonstrated the potential interest of the proposed system

Mixed-Reality Prototypes to Support Early Creative Design

The domain we address is creative design, mainly architecture. Rooted in a multidisciplinary approach as well as a deep understanding of architecture and design, our method aims at proposing adapted mixed-reality solutions to support two crucial activities: sketch-based preliminary design and distant synchronous collaboration in design. This chapter provides a summary of our work on a mixed-reality device, based on a drawing table (the Virtual Desktop), designed specifically to address real-life/business-focused issues. We explain our methodology, describe the two supported activities and the related users’ needs, detail the technological solution we have developed, and present the main results of multiple evaluation sessions. We conclude with a discussion of the usefulness of a profession-centered methodology and the relevance of mixed reality to support creative design activities.