《A Database Architecture For Real-Time Motion Retrieval》.pdf

2009 Seventh International Workshop on Content-Based Multimedia Indexing A Database Architecture For Real-Time Motion Retrieval Charly Awad, Nicolas Courty and Sylvie Gibet VALORIA, University of Bretagne-Sud Vannes, FRANCE charly.awad nicolas.courty sylvie.gibet@univ-ubs.fr Abstract Over the past decade, many research fields have real- ized the benefits of motion capture data, leading to an ex- ponential growth of the size of motion databases. Conse- quently indexing, querying, and retrieving motion capture data have become important considerations in the usability of such databases. Our aim is to efficiently retrieve mo- tion from such databases in order to produce real-time an- imation. For that purpose, we propose a new database ar- chitecture which structures both the semantic and raw data contained in motion data. The performance of the overall architecture is evaluated by measuring the efficiency of the motion retrieval process, in terms of the mean time access to the data. Figure 1. Creating animation from motion chunks extracted from the database. 1 Introduction With the development of new technologies for motion capture, databases of human motions have become more A key aspect of motion data is that it contains multiple accessible and yielded new methods for motion indexing, levels of information, from the semantic level to the time and retrieval, and computer-generated animation. As these