报告题目：Tree-based shape spaces: definition and applications in image processing and computer vision 报告人：许永超博士 时  间：6月 22日  09:30 地  点：行政楼会议室108

报告简介：

Connected operators are filtering tools that act by merging elementary regions of an image. A popular strategy is based on tree-based image representations: for example, one can compute a shape-based attribute on each node (connected component) of the tree and keep only the nodes for which the attribute is sufficiently strong. This operation can be seen as a thresholding of the tree, seen as a graph whose nodes are weighted by the attribute. We call this graph of connected components the tree-based shape space, which features several interesting properties. An example showing its usefulness is the proposed Tree-Based Morse Regions (TBMR), a local feature detector that is fully invariant to increasing change of contrast. Rather than being satisfied with a mere thresholding of the shape space, we propose to expand on the idea of connected operators by applying connected filters on the shape space. Such a processing is a generalization of the existing tree-based connected operators. Indeed, the framework includes classical existing connected operators by attributes. It also allows us to propose a class of novel connected operators. Moreover, this framework can also be used for object segmentation and hierarchical image segmentation.

个人简介：

Yongchao Xu received the BSc degree in Optoelectronics from Huazhong University of Science and Technology in 2008, the engineer degree in electronics & embedded systems from Polytech Paris Sud and the master degree in signal processing & image processing from Université Paris Sud 11 in 2010, and the Ph.D. degree in image processing and mathematical morphology from Université Paris Est in 2013. He is currently working at EPITA Research and Development Laboratory (LRDE) as an assistant professor, and at the Signal and Image Processing Department, Telecom ParisTech as a visiting profesor. His research interests include mathematical morphology, image segmentation, text detection, medical image analysis, and local feature detection.