题 目（TITLE）：Detecting and Segmenting Objects in Images
讲 座 人（SPEAKER）: Prof. Jitendra Malik；University of California at Berkeley
主 持 人 (CHAIR)：Associate Prof. Huang Kaiqi
时 间 (TIME)： 10:30AM, October 21 (Tuesday)
地 点 (VENUE)： The Second Meeting Room, 13th Floor
Detection, segmentation and 3D pose extraction of people in images are challenging problems in visual recognition. Existing methods can only solve these problems partially. We propose the notion of "poselets", parts that are tightly clustered in the appearance space of image patches and the configuration space of keypoints. Individual poselet activations are noisy, but the spatial context of each can provide vital disambiguating information. This can be done by training a two-layer feed-forward network. The refined poselet activations are then clustered into mutually consistent hypotheses where consistency is based on empirically determined spatial keypoint distributions. Finally, bounding boxes are predicted for each person hypothesis and shape masks are aligned to edges in the image to provide a segmentation. Our system performs best on people detection and segmentation. These ideas extend naturally to other visual categories, and offer a unified theory of object detection and segmentation.Joint work with Lubomir Bourdev, Subhransu Maji and Thomas Brox.
Jitendra Malik received the B.Tech degree in Electrical Engineering from Indian Institute of Technology, Kanpur in 1980 and the PhD degree in Computer Science from Stanford University in 1985. In January 1986, he joined the University of California at Berkeley, where he is currently the Arthur J. Chick Professor in Department of Electrical Engg and Computer Sciences.During 2002-2004 he served as the Chair of the Computer Science Division and during 2004-2006 as the Department Chair of EECS. He serves on the advisory board of Microsoft Research India, and on the Governing Body of IIIT Bangalore. He was awarded the Longuet-Higgins Prize in 2007 and in 2008. He is a fellow of the IEEE and the ACM.
His current research interests are in computer vision, computational modeling of human vision and analysis of biological images. His work is in vision including image segmentation, perceptual grouping, texture, stereopsis and object recognition with applications to image based modeling and rendering in computer graphics, intelligent vehicle highway systems, and biological image analysis.