Active learning for real-time motion controllers

Seth Cooper, Aaron Hertzmann, Zoran Popović
2007 ACM SIGGRAPH 2007 papers on - SIGGRAPH '07  
Figure 1 : Catching controller: the character moves through space to catch two consecutive balls thrown from different directions. In addition to the current character state, the controller has a three-dimensional input space that specifies the incoming position and speed of the ball to be caught. Abstract This paper describes an approach to building real-time highlycontrollable characters. A kinematic character controller is built on-the-fly during a capture session, and updated after each new
more » ... motion clip is acquired. Active learning is used to identify which motion sequence the user should perform next, in order to improve the quality and responsiveness of the controller. Because motion clips are selected adaptively, we avoid the difficulty of manually determining which ones to capture, and can build complex controllers from scratch while significantly reducing the number of necessary motion samples.
doi:10.1145/1275808.1276384 fatcat:4ccz567lcnahna2kelbtsl4ste