Plenary Sessions

Title: Computer Vision:  The Next Decade

Speaker: Prof. Larry Davis


Larry S. Davis received his B.A. from Colgate University in 1970 and his M. S. and Ph. D. in Computer Science from the University of Maryland in 1974 and 1976 respectively. From 1977-1981 he was an Assistant Professor in the Department of Computer Science at the University of Texas, Austin. He returned to the University of Maryland as an Associate Professor in 1981. From 1985-1994 he was the Director of the University of Maryland Institute for Advanced Computer Studies. He was Chair of the Department of Computer Science from 1999-2012. He is currently a Professor in the Institute and the Computer Science Department, as well as Director of the Center for Automation Research.   He was named a Fellow of the IEEE in 1997 and of the ACM in 2013.

Prof. Davis is known for his research in computer vision and high performance computing. He has published over 100 papers in journals and 200 conference papers and has supervised over 40 Ph. D. students.  During the past ten years his research has focused on visual surveillance and general video analysis. He has served as program or general chair for most of the field's major conferences, including the 5’th International Conference on Computer Vision, the 2004 Computer Vision and Pattern Recognition Conference, the 11’th International Conference on Computer Vision held in 2006, the 2010 Computer Vision and Pattern Recognition Conference, and the 2013 International Conference on Computer Vision .

 

 Abstract The field of computer vision has advanced remarkably during the past 10-15 years.  This is due to a variety of factors including the availability of the large annotated data sets needed to train deep learning models, software like AMT that enables the collection of these data sets at reasonable costs, important engineering improvements to  the training methodologies of deep networks, dramatic decreases in price/performance ratios of computing systems (especially GPU’s) and memory systems, widespread availability of source code that researchers make available to one another worldwide, and inexpensive sensors and robotic platforms like Kinect, Go-pro’s and UAV’s .  So, while the fundamental vision problems of detection and recognition of objects and human movements are not solved, they have improved to the point where it is important to ask: What’s next?  A workshop was held in the US late last year to address  exactly that question (chaired by me, Fei Fei Li and Devi Parikh) and this talk will discuss the conclusions of that workshop, and illustrate research in some of those future directions with work from the University of Maryland, in particular research on visual search .  I will describe a general strategy for object detection, that instead of passively evaluating all object detectors at all possible locations in an image, employs  a divide-and-conquer approach by actively and sequentially evaluating contextual cues related to the query based on the scene and previous evaluations—like playing a “20 Questions” game—to decide where to search for the object. The problem is formulated as a Markov Decision Process and a search policy is learned by reinforcement learning. To demonstrate the efficacy of the algorithm, it is applied to the 20 questions approach in the recent framework of simultaneous object detection and segmentation.

 

Title: Design of Color Filter Arrays                         

Speaker: Zhouchen Lin                                                    


     
Zhouchen Lin received the Ph.D. degree in applied mathematics from Peking University in 2000. He is currently a Professor at Key Laboratory of Machine Perception (MOE), School of Electronics Engineering and Computer Science, Peking University. He was a Chair Professor at Northeast Normal University, a guest professor at Beijing Jiaotong University, Shanghai Jiao Tong University and Southeast University, and a guest researcher at Institute of Computing Technology, Chinese Academy of Sciences. Before March 2012, he was a Lead Researcher at Visual Computing Group, Microsoft Research Asia. His research interests include computer vision, image processing, computer graphics, machine learning, pattern recognition, and numerical computation and optimization. He is an associate editor of IEEE T. Pattern Analysis and Machine Intelligence and International J. Computer Vision and an area chair of CVPR 2014, ICCV 2015, NIPS 2015, AAAI 2016, CVPR 2016, and IJCAI 2016.

 

Abstract: A color filter array (CFA) is a mosaic of spectrally selective filters used in a digital camera, which allows only one color component to be sensed at each pixel. The missing two components of each pixel have to be estimated by methods known as demosaicking. The demosaicking algorithm and the CFA design are crucial for the quality of the output color images. In this talk, I will introduce my recent work on designing CFAs and show the superior performance of newly found CFAs.