【講座題目】 3D Point Match Evaluation Through Adapting The Gentle AdaBoost

【講座時(shí)間】 2017年7月5日(周三) 上午：9:00-11:00

【講座地點(diǎn)】 保定校區(qū)教十樓A612

【主 講 人】 Yonghuai Liu，副教授、英國(guó)亞伯大學(xué)

【主講人簡(jiǎn)介】

  Yonghuai Liu，博士，IEEE高級(jí)會(huì)員，在圖像處理、計(jì)算機(jī)視覺(jué)、機(jī)器人路徑規(guī)劃、智能信息處理、進(jìn)化計(jì)算等領(lǐng)域頗有建樹(shù)，是《Neurocomputing》、《Pattern Recognition Letters: an international journal》、《American Journal of Educational Research》等國(guó)際知名刊物的副主編和編委，同時(shí)也是2014、2015 IEEE Interntional Conference on Robotics and Automation國(guó)際會(huì)議的副主席。迄今為止，在德國(guó)斯普林格（Springer-Verlag）等出版社出版著作4部，公開(kāi)發(fā)表高水平學(xué)術(shù)論文140余篇，主持英國(guó)和國(guó)際合作科技項(xiàng)目10余項(xiàng)，其成果成功應(yīng)用于工業(yè)檢測(cè)、工業(yè)機(jī)器人等行業(yè)。

【報(bào)告內(nèi)容簡(jiǎn)介】

   3D partial shapes nowadays can be easily captured using various kinds of systems such as Microsoft Kinect and structure from motion. While such systems have limited field of view, it is necessary to capture multiple datasets from different positions in order to cover a large area of the surface of the object of interest. To fuse information from these datasets, they have to be aligned into the same coordinate system. This process is called registration and the feature extraction and matching (FEM). FEM finds numerous applications in the real world for object modeling, industrial quality assurance, and computer games and virtual reality. In this talk, I will firstly show the data formats and their visualization, then discuss the issues for the feature extraction and representation and outline two important techniques for the task.

                                                                                  科學(xué)技術(shù)處   計(jì)算機(jī)系
                                                                                    二〇一七 年 六月二十九日