关于举办“Structured Learning and Its Applications in Computer Vision”学术报告通知

报告时间：2014年11月24日16：00到17:00

报告地点：det365官方网站登录副203# 

报告内容：

Structured Learning is an important branch in Statistical Learning. It has widely been applied to a number of computer vision tasks such as human action recognition, image segmentation, object detection and scene understanding. The talk first briefly introduces probably the most popular structured learning techniques including structured SVM and latent structured SVM. After this, the talk concentrates on the challenging inference problems in structured learning, especially the Markov random field (MRF) inference with unknown graphs. Two novel inference algorithms will be discussed. Evaluations made on semantic image segmentation and human action recognition show the significance of inferring graph structures and the improvement made by our algorithms compared with the state-of-the-art.

报告人简介：

王振华，男，河南省开封县人。2003年9月至2010年7月就读于det365官方网站登录，分别获工学学士和硕士学位。2010年9月获国家留学基金委（CSC）资助，公派至澳大利亚阿德莱德大学（The University of Adelaide）攻读博士学位，2014年9月提交的博士学位论文，已通过专家评审。博士期间研究课题是机器学习和计算机视觉，主攻未知拓扑结构的概率图推理算法及其应用的研究。读博士期间在国际顶级会议和期刊上（包括CVPR，ACCV，TPAMI）发表三篇论文，其中一篇论文荣获ACCV 2014 最佳论文奖，另有两篇论文在审。