报告题目：A Comparative Study of Image Feature Detectors and Descriptors
报告人：Jonathan Wu, University of Windsor, Canada
时间： 4月23号 10：00-12：00
地点：CSC-223

Abstract ：Feature detection and matching is a fundamental problem in many computer and robot vision applications. During the past decades, various types of feature detectors and descriptors have been proposed in the literature.
This presentation provides a comparative study of a large number of popular feature detectors from last three decades. The study makes several contributions towards a generic comparison of feature detectors and descriptors. Firstly, comparisons on invariance against image transformations such as illumination changes, blurring, rotation, scaling, viewpoint changes, exposure, JPEG compression, combined scaling and rotation, combined viewpoint changes, scaling, and rotation are conducted. The second contribution of the work is to provide a proper distinction between detectors and descriptors with separate comparisons. Exhaustive experiments on several datasets for each combination of detector and descriptor have been conducted to provide a ranking which can also be weighted to suit specific applications.

报告人简介：

    武庆明（Q.M. Jonathan Wu）教授1990年毕业于英国威尔士-斯旺西大学，获得电子工程博士学位。1995年他加入加拿大国家研究院(National Research Council of Canada)，并成为高级研究主管。2005年起任教加拿大温莎大学University of Windsor，现任计算机视觉和传感系统研究室主任，加拿大汽车电子和信息系统的首席科学家。
    武庆明教授主要从事计算机和机器人视觉，图像处理， 模式识别和智能系统的研究。武教授已在相关国际学术会议及国际核心期刊上发表学术论文300余篇，其中包括150多篇的国际期刊论文 (50多篇为IEEE Transaction论文)。武教授曾担任五十多个国际会议的组织委员会主席和评审委员会委员，现任IEEE Transaction on Neural Networks and Learning Systems, IEEE Transaction on Systems, Man, and Cybernetics-Part A 与International Journal of Robotics and Automation的副主编。