时 间：2012年12月6日星期四下午2：30-3：30
地 点：浙江大学紫金港校区图书信息中心B楼CAD&CG国家重点实验室402室
报告题目：Visual content abstraction based on high sparsity and variance measures
报告人：贾佳亚(香港中文大学副教授)
主持人：章国锋 副教授
Abstract:
I will introduce two new visual content abstraction tools. One is effective for sharpening major edges by increasing the steepness of transitions while eliminating a manageable degree of low-amplitude structures. The seemingly contradictive effect is achieved in an unconventional optimization framework making use of L0 gradient minimization, which can globally control how many non-zero gradients are resulted in to approximate prominent structures in a structure-sparsity-management manner. The other tool can quickly remove texture from images without extensive prior knowledge. It is also based on global optimization and uses a new natural variation constraint. These two methods are fundamental, find a wide spectrum of applications and vastly benefit computer vision, signal processing, graphics and multimedia in, for example, signal separation, segmentation, abstraction, edge extraction,clip-art JPEG artifact removal, and non-photorealistic image generation.
Bio:
Jiaya Jia is currently an associate professor in the Department of Computer Science and Engineering at The Chinese University of Hong Kong (CUHK). He was a visiting scholar at Microsoft Research Asia from March 2004 to August 2005 and conducted collaborative research at Adobe Systems in 2007. He leads the research group at CUHK focusing specifically on computational photography, 3D reconstruction, practical optimization, and motion estimation. He serves as an associate editor for the IEEE Transactions on Pattern Analysis and Machine Intelligence and served (or will serve) as an area chair for ICCV 2011 and ICCV 2013. He was on the program committees of several major conferences, including ICCV, ECCV, and CVPR, and co-chaired the Workshop on Interactive Computer Vision, in conjunction with ICCV 2007. He won the best paper award in NPAR 2012. He is a senior member of the IEEE.
