On one hand, humans perceive visual media through their innate human visual system (HVS), known to have unique characteristics and limitations (limited speed and types of eye movements, inability to discern details away from gaze focal point, etc). On the other hand, visual media (e.g., images and video) contains low-level salient features like motion and flickers that draw human's visual attention. In this research, we seek to derive useful semantic information by jointly analyzing physical signals emitted by observer's HVS and salient features inherent in the visual media. In particular, I will present first a gaze-based video streaming system, where the future gaze location of an observer is predicted based on previous tracked gaze patterns and pre-computed visual saliency maps of the observed video, so that only the spatial regions of future video frames are encoded in high quality, reducing streaming rate. I will discuss an extension of this streaming system, where streaming rate can be effectively reduced even for store-and-playback video content with no real-time encoding. Finally, I will discuss how one can classify video into "quiet" and "busy" according to the amount of attention shifts a video clip will induce an observer. Because of frequent attention shifts, gaze prediction for "busy" video is much harder.

Biography

Gene Cheung received the B.S. degree in electrical engineering from Cornell University in 1995, and the M.S. and Ph.D. degrees in electrical engineering and computer science from the University of California, Berkeley, in 1998 and 2000, respectively. He was a senior researcher in Hewlett-Packard Laboratories Japan, Tokyo, from 2000 till 2009. He is currently an assistant professor in National Institute of Informatics in Tokyo, Japan.

His research interests include robust media representation, single- / multiple-view video coding & streaming, and immersive communication. He has published over 100 international conference and journal publications. He has served as associate editor for IEEE Transactions on Multimedia from 2007 to 2011 and currently serves as associate editor for DSP Applications Column in IEEE Signal Processing Magazine and APSIPA journal on signal and information processing. He has also served as area chair in IEEE International Conference on Image Processing (ICIP) 2010 and 2012, technical program co-chair of International Packet Video Workshop (PV) 2010 and track co-chair for Multimedia Signal Processing track in IEEE International Conference on Multimedia and Expo (ICME) 2011. He serves as symposium co-chair for CSSMA Symposium in IEEE GLOBECOM 2012.

He is a co-author of top 10% paper in IEEE International Workshop on Multimedia Signal Processing (MMSP) 2009 and 2011, best student paper in IEEE Workshop on Streaming and Media Communications 2011 (in conjunction with ICME 2011) and best paper finalists in ICME 2011 and ICIP 2011.