Human motion reconstruction from sparse 3D motion sensors using kernel CCA-based regression

This paper presents a real-time performance animation system that reproduces fullbody character animation based on sparse 3Dmotion sensors on a performer. Producing 1 In Computer Animation and Virtual Worlds, 24(6), 565-576, 2013 This version is the author's manuscript. faithful character animation from this setting is a mathematically ill-posed problem because input data from the sensors are not sufficient to determine the full degrees of freedom of a character. Given the input data from 3D motion sensors, we select similar poses from a motion database and build an online local model that transforms the low-dimensional input signal into a high-dimensional character pose. A regression method based on kernel CCA (Canonical Correlation Analysis) is employed because it effectively handles a wide variety of motions. Examples show that various human motions are naturally reproduced by the proposed method.