报告题目：Introduction of image Co-segmentation and recent works
报告摘要：Image segmentation is the fundamental question in computer vision society, which plays important roles in many senior tasks and receives broad attention. Recently, image co-segmentation has become a new hot area in segmentation. The main task of co-segmentation is to mine the consistencies and connections of the images in a group, which will then assist the segmentation of each image. In this talk, I will give a brief review of image co-segmentation and introduce the new challenges in this field. Besides, we will introduce our recent works on this topic.
 Kunqian Li, Jiaojiao Zhang and Wenbing Tao*. Unsupervised Co-Segmentation for Indefinite Number of Common Foreground Objects, IEEE Transactions on Image Processing, 25(4): 1898-1909, 2016.
 Wenbing Tao*, Kunqian Li, Kun Sun. SaCoseg: Object Cosegmentation by Shape Conformability. IEEE Transactions on Image Processing, 24(3): 943-955, 2015.
 Kunqian Li, Wenbing Tao*. Adaptive Optimal Shape Prior for Easy Interactive Object Segmentation, IEEE Transactions on Multimedia. 17(7): 994-1005, 2015.
 Jiaojiao Zhang, Kunqian Li* and Wenbing Tao. Multi-video Object Cosegmentation for Irrelevant Frames Involved Videos, IEEE Signal Processing Letter, 23(6): 785-789, 2016.
 Liman Liu, Kunqian Li* and Xiangli Liao. Image Co-segmentation by Co-diffusion, Circuits, Systems, and Signal Processing, in print.
Kunqian Li is currently pursuing the Ph.D. degree with the National Key Laboratory of Science and Technology on Multi-Spectral Information Processing, School of Automation, Huazhong University of Science and Technology, Wuhan, China, with Prof. Wenbing Tao. His research interests include image segmentation and object recognition, the related research results have been published in authoritative journals, such as IEEE TIP, TMM, etc. He received National Scholarship for Graduate Students in 2016.
GMT+8, 2017-3-30 06:49 , Processed in 0.040738 second(s), 19 queries .