Computational photography is pushing sophisticated computational thinking deeper into the imaging pipeline. For example, light field cameras record higher-dimensional data than conventional cameras, and enable new functionality, like depth inference, refocusing and correcting lens aberrations in post-processing. This talk will review the theory and intuition of light field camera design, and describe the resulting transformation of design considerations and opportunities in core subsystems: optics, sensors and processors. The development of such cameras opens the door to the use of light field cameras for computer vision and machine learning.
|Ren Ng is a faculty member in Electrical Engineering and Computer Science at the University of California, Berkeley. His research interests are in imaging, graphics and applied mathematics, focusing on the theory and engineering of computational imaging systems. In 2006, Ren founded Lytro, Inc., which commercialized his Ph.D. research and brought consumer light field cameras to market. Ren completed his Ph.D. in computer science at Stanford University. Awards include the ACM Doctoral Dissertation Award, Sloan Research Fellowship, Selwyn Medal from the Royal Photographic Society, MIT Tech Review's TR35 and Entrepreneur of the Year, Fast Company's 100 Most Creative People in Business, and Silicon Valley Journal's 40 under 40.|
Although realistic face/hair modeling and animation technologies have been widely employed in computer generated movies, it remains challenging to deploy them in consumer-level applications such as computer games, social networks and other interactive applications. The main difficulties come from the requirement of special equipment, sensitivity to daily environments, laborious manual work and high computational costs. In this talk, I will introduce our recent research on realistic face/hair modeling and animation, aiming at interactive applications and ordinary users. In particular, I will describe fully automatic approaches to real-time facial tracking and animation with a single web camera, methods for modeling hairs from images, and real-time algorithms for realistic hair simulation.
|Kun Zhou is a Cheung Kong Professor and the Director of the State Key Lab of CAD&CG at Zhejiang University. Earlier, he was a Lead Researcher of the Internet Graphics Group at Microsoft Research Asia. He received his BS and PhD degrees from Zhejiang University in 1997 and 2002, respectively. His research interests include geometry processing, photorealistic rendering, computer animation and GPU parallel computing. He is/was an associate editor of IEEE TVCG and ACM TOG, and serves on the editorial advisory board of IEEE Spectrum. He is a Fellow of IEEE.|