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.
Curved Reflection Symmetric Axes on Free-form Surfaces and Their Extraction
Feature detection on smooth free-form surfaces is much more difficult than that on shapes with sharp features. In this paper, we extract the curved reflection axes of an arbitrary free-form surface as features if they exist. The extraction result is robust to boundary noises and strongly sensitive to the extrinsic properties of the surface such as projected normals and curvatures. Compared with the general reflection symmetry, curved reflection symmetry is defined to be a reflection symmetry along a smooth 3D embedded curve instead of a plane, where any point on the curve is a local reflection center for some surface points. The properties of the curved reflection symmetric axis are analyzed, and a novel computational model for detecting and extracting curved reflection axes (CRA) on free-form surfaces is presented. The experimental results are then compared with both the medial axis and the intrinsic symmetric axis, which are two popular feature representations of 3D shapes, and the advantages and uniqueness of the proposed method are convincingly demonstrated. An application of the proposed method in sweep scanning is also presented. Other applications of the proposed method include feature extraction, shape symmetrization, segmentation and registration, etc.
从1991年六月至2001年六月，整十年，汤凯就职于美国 APPLICON 公司 （后被 UniGraphics 收购），主要从事 CAD/CAM 软件的开发和研制，任高级软件专家。在此其间，利用业余时间，在 CAD，GRAPHICS MODELS AND IMAGE PROCESSING， ASME JOURNAL OF MECHANICAL DESIGN， COMPUTERS IN INDUSTRY 上发表了十几篇论文。
自2001年七月起，汤凯被香港科技大学聘为助理教授，从事教学和科研工作，并于2011年提升为正教授。自从受聘于香港科大，至今十四年间，作为首席科学家，汤凯获得众多香港政府自然基金和科技创新基金研究项目，总额超过港币两千六百万，并与诸如中国航空集团、通用电气等大公司建立了紧密的研究伙伴关系。同时，他还和中国诸多大学联系紧密，如受聘西北工业大学讲座教授、南京航天航空大学客座教授等。汤凯目前的研究方向为：几何造型，曲面平展算法，五轴加工路线的产生及优化算法，快速成型加工的优化算法，神经网络理论在机械设计及制造中的应用，现为《Computer-Aided Design》、《Journal of Computational Design and Engineering》等国际杂志的编委。