Invited Talk by Dr. Norimichi Ukita: 3D human pose tracking using action-specific motion and regression models

Dr. Norimichi UKITA Date: Monday, March 4, 2013 Time: 14:00 - 14:30 Location: Seminar room MI 03.13.054

3D human pose tracking using action-specific motion and regression models

Abstract: We propose a unified model for human motion prior with multiple actions. Our model is generated from sample pose sequences of the multiple actions, each of which is recorded from real human motion. The sample sequences are connected to each other by synthesizing a variety of possible transitions among the different actions. For kinematically-realistic transitions, our model integrates nonlinear probabilistic latent modeling of the samples and interpolation-based synthesis of the transition paths. While naive interpolation makes unexpected poses, our model rejects them 1) by searching for smooth and short transition paths by employing the good properties of the observation and latent spaces and 2) by avoiding using samples that unexpectedly synthesize the non-smooth interpolation. The effectiveness of the model is demonstrated with real data and its application to human pose tracking.

Bio: Norimichi Ukita is an associate professor at the graduate school of Information Science, Nara Institute of Science and Technology (NAIST), Japan. He received the Ph.D degree in Informatics from Kyoto University, Japan, in 2001 under the supervision of Dr. Takashi Matsuyama. After working for five years as an assistant professor at NAIST, he became an associate professor in 2007. He was a research scientist of Science and Technology Agency (JST), during 2002 - 2006. He was a visiting research scientist at Carnegie Mellon University working with Dr. Takeo Kanade, during 2007-2009. He is also currently a visiting researcher at Advanced Telecommunications Research Institute Internatioal (ATR).