“Motion belts” by Yasuda, Kaihara, Saito and Nakajima

Large scale motion databases have became essential for both computer animation and human motion analysis. We propose a novel visualization style of motion clip’s outlines that unfolds a motion clip into a 2D stripe of keyframes. We call it a motion belt. Because of the form factor of the stripes, motion belts are suitable to display the outlines of the multiple motions simultaneously. This enables users to compare the multiple clips instantaneously skimming through the motion database. The existing outline visualization methods for motions extract keyframes and place them in a 3D space [Assa et al. 2005]. However, for the long clips without large locomotion, boxing for instance, the visualization results in a large cloud of keyframes that is hard to recognize the posture of each keyframe and the order of the actions in the clip.

1. Assa, J., Caspi, Y., and Cohen-Or, D. 2005. Action synopsis: pose selection and illustration. ACM Trans. Graph. 24, 3, 667–676.
2. Vazquez, P. P., Feixas, M., Sbert, M., and Heidrich, W. 2001. Viewpoint selection using viewpoint entropy. In Proceedings of the Vision, Modeling, and Visualization (VMV) Conference, 273–280.