The development of methods and tools for the generation of visually appealing motion sequences using prerecorded motion capture data has become an important research area in computer animation. In particular, data-driven approaches have been used for reconstructing high-dimensional motion sequences from low-dimensional control signals. In this article, we contribute to this strand of research by introducing a novel framework for generating full-body animations controlled by only four 3D accelerometers that are attached to the extremities of a human actor. Our approach relies on a knowledge base that consists of a large number of motion clips obtained from marker-based motion capturing. Based on the sparse accelerometer input a cross-domain retrieval procedure is applied to build up a lazy neighborhood graph in an online fashion. This graph structure points to suitable motion fragments in the knowledge base, which are then used in the reconstruction step. Supported by a kd-tree index structure, our procedure scales to even large datasets consisting of millions of frames. Our combined approach allows for reconstructing visually plausible continuous motion streams, even J. Tautges and T. Helten were financially supported by grants from Deutsche Forschungsgemeinschaft (WE 1945/5-1 and MU 2686/3-1).

No abstract

Abstract:We investigate the use of multi-linear models to represent human motion data. We show that naturally occurring modes in several classes of motion can be used to efficiently represent the motions for various animation tasks, such as dimensionality reduction or synthesis of new motions by morphing. We show that especially for the approximations of motions by few components the reduction based on a multi-linear model can be considerably better than one obtained by principal component analysis (PCA).