Ming-Ming Cheng


Ming-Ming Cheng is a professor with CCCE, Nankai University. He received his PhD degree from Tsinghua University in 2012. Then he worked as a research fellow for 2 years, working with Prof. Philip Torr in Oxford. Dr. Cheng’s research primarily centers on algorithmic issues in image understanding and processing, including image segmentation, editing, retrieval, etc. He has published over 30 papers in leading journals and conferences, such as IEEE TPAMI, ACM TOG, ACM SIGGRAPH, IEEE CVPR, and IEEE ICCV. He has designed a series of popular methods and novel systems, indicated by 6000+ paper citations (2000+ citations to his first author paper on salient object detection). His work has been reported by several famous international media, such as BBC, UK telegraph, Der Spiegel, and Huffington Post.   [CV]


I’m currently working on image scene analysis, editing, and retrieval. These works are mainly in following aspects: (I) biological motivated salient region detection and segmentation; (II) sketch based image retrieval and composition; (III) interactive image analysis and manipulation; (IV) similar scene elements analysis for smart image manipulation. These works tried to recover parts of scene object level information from images according to biological inspiration, or with the help of simple user assist in sketch form. Such scene object level information includes one or more parts of the following aspects: object of interest regions, object correspondence, region layering, symmetry, repetition, and 3D relations. Ideally, we expect automatic extraction of full 3D information, categories names, attributes, and object relations about the underlying image scene for intelligent image understanding, manipulation, organization, and retrieval. [Research galleries]


My publications could be found here. See also: DBLPGoogleScopusArnetminer, ResearchID, arXiv

Major Honors & Awards


Locations of visitors to this page


12 Comments on "Ming-Ming Cheng"

Notify of
Runxin Ma
Runxin Ma

您好,我从您的网站上下载了SED2_ImgGt图库资源,请问为什么SED2_GT(ground truth)中的的图像不是二值图(不是单纯有0和255这两地灰度值),这是原始的图库数据吗?