【電機系書報討論演講訊息】

書報討論網址：http://www.ee.nthu.edu.tw/~ee591000/

演講公告網址：/p/403-1175-2629-1.php

題   目：Emerging Video Technologies for Resolution Explosion and Beyond - H.265/HEVC and Light-Field Camera

講   員：黃朝宗 博士

時       間：2012年12月24日(星期一) 上午10時至11時

地       點：資電館206教室
 

Abstract

Video resolution is now blowing up for digital devices from cellphone to television, even beyond human perceptual limits. In this talk, I will introduce two emerging video technologies which can help continue this trend in short-term and long-term future respectively.

The first is the newest video coding standard, H.265/HEVC, which achieves twice of compression capability of H.264. It can alleviate the transmission and storage penalties brought by the skyrocketing resolutions, however, at the expense of more complexity and memory. I will focus on the evolution from previous standards and the design challenges for real-time systems compared to H.264.

The second is the light-field camera which transfers the resolution beyond perceptibility into useful angular one. The lens-array camera which I am working on will be introduced. It shows the opportunity that we could not only take SLR-quality pictures with a body as thin as cellphones but also provide digital refocus, scalable depth-of-field and autostereoscopic 3D images, all in digital processing. I will focus on the design principles and challenges as well as possible extensions in this era of resolution explosion.

Biography

Chao-Tsung Huang received the B.S. and PhD degrees in electrical engineering both from National Taiwan University in 2001 and 2005 respectively. He is now performing postdoctoral research on lens-array light-field camera at National Taiwan University.

He worked for Novatek Microelectronics from 2005 to 2011 as an IC design team leader for developing multi-standard video codecs, with fifteen patents filed. During 2011-2012, he was a postdoctoral associate at Massachusetts Institute of Technology, Cambridge, and had developed the world-first H.265/HEVC decoder chip.

He has authored and co-authored one book, eleven journal papers, and nineteen conference papers. He received the MediaTek Fellowship from 2003 to 2005. His research interests are high efficiency video codec and light-field processing, from algorithm exploration to VLSI design, chip implementation, and demo system.

歡  迎  聽  講