Polynomial Local Shape Descriptor on Interest Points
For 3D Part-in-whole Matching
Part-in-whole 3D shape matching is to recognize query shapes as sub-parts of a target intact 3D object. It plays a pivotal role in a large number of engineering applications. The most critical component in a part-in-whole search system is the local shape descriptor which encapsulates the identified local feature on the query part and is matched with the local shape descriptors of the parts in the database. We propose a novel local shape descriptor based on the concept that the evolution pattern of geodesic iso-contour’s length is a good representative for surface features. Our local shape descriptor enjoys a unique advantage over most existing ones by being sensitive to the geodesic radius of the local region, and thus is able to capture more comprehensive shape information if the query portion of the shape is larger and includes more complicated surface features. Through a simple approximation scheme, our local shape descriptor is defined as a vector piecewise polynomial function of the geodesic radius of the interest point, thus enabling local matching to be performed quickly by simple curve evaluations. We also introduce a new schema of interest points sampling so that we can reserve the most corresponding information of the model by a small number of local feature descriptors. The proposed part-in-whole matching approach outperforms many existing approaches in matching efficiency and requiring a smaller input region. It is a shortcut solution for incomplete model matching/retrieval.
About the Speaker：
Dr. Kai Tang received BSE in Mechanical Engineering from Nanjing Institute of Technology (now Southeast University) in 1982 and afterward, as sponsored by the Chinese government, went to the University of Michigan (Ann Arbor) where he received PhD in Computer Engineering in 1990. From 1991 to 2001, he worked as a software specialist at Schlumberger CAD/CAM (which in 1994 became Applicon and in 1999 became part of Unigraphics). Among his various responsibilities and projects, Dr. Kai Tang was one of the chief architects of the multi-axis NC surface machining software BravoNC; he also led a team that developed an ACIS-based geometric engine for 5-axis tool path computation. In 2000, he also for a short period of one year worked as the chief engineer in a start-up company focusing on voice recognition technology. In June 2001, he “went back” to school and joined the faculty of the Department of Mechanical and Aerospace Engineering at Hong Kong University of Science and Technology (HKUST), arising from Assistant professor in 2001, Associate professor in 2006, and Full professor in 2011. Dr. Kai Tang’s research interests are broad, but mostly concentrating on developing efficient and practical algorithms in CAD/CAM. In his less than 14 years with HKUST, as sole PI he has obtained in total of HK$26 million in research funding from the Hong Kong RGC General Research Fund and Innovative Technology Fund. Aside from his regular professional research and teaching in academia, Dr. Kai Tang is also an avid writer http://ihome.ust.hk/~mektang/public_files/here_and_there.html.