报告人:Dr. Yunchao Wei



报告主题Towards Weakly- and Semi- Supervised Object Localization and Semantic            Segmentation


摘要:Over the past few year, the great success of CNNs in object detection and image semantic segmentation relies on a large amount of human annotations. However, collecting annotations such as bounding boxes and segmentation masks is very costly. To relieve the demand of finance and human effort, in this talk, Dr. Yunchao Wei will introduce his recent works, which utilize weak information as supervision to address more challenging object localization and semantic segmentation tasks. In particular, he proposes several novel solutions to produce dense object localization maps only using image-level labels as supervision. The dense object localization maps can successfully build the relationship between image-level labels and pixels, and effectively boost the accuracy of localization and segmentation tasks. His works are published on top-tier journals/conferences (e.g. T-PAMI and CVPR) and achieve state-of-the-art performance. 


报告人简介:魏云超博士, 2016年获得北京交通大学(BJTU)信号与信息处理专业博士学位,导师为赵耀教授;2013年到2017年之间在新加坡国立大学(NUS)跟随颜水成博士和冯佳时博士从事访问学者和博士后的研究工作;现为美国伊利诺伊大学香槟分校(UIUC)四院院士Thomas Huang教授的博士后研究员;迄今发表学术论文30余篇,其中包括以下顶级期刊/会议:T-PAMI (4), T-IP, T-CSVT, T-CYB, T-MM, T-NNLS, TIST, PR, CVPR (7), ICCV, AAAI, MM,Google学术引用次数730+。魏云超博士曾获得2016年中国电子学会和北京交通大学优秀博士论文奖;曾获得ImageNet视觉挑战赛在图像物体检测任务(ILSVRC-2014)和视频物体检测任务(ILSVRC-2017)的冠亚军;曾担任T-PAMI, CVPR等本领域主流期刊/会议的审稿人。其研究领域涉及计算机视觉和多媒体分析,主要包括:多标签分类,物体检测,语义分割,弱监督/半监督学习,多模态数据分析等。