Abstract: 3D shapes have been applied in many fields, such as industrial design, education, biological medicine, animation entertainment and preservation of cultural relics, etc. Feature extraction of 3D shapes is an important branch of computer graphics and pattern recognition, attracting more attention from many researchers of late years. In particular, the deformations of non-rigid 3D shapes, which mostly contain articulated structure, increase the difficulty of feature extraction of 3D shapes. This paper focuses on feature extraction algorithms of rigid 3D shapes and non-rigid 3D shapes, which are proposed in recent years, analyzes the difficulties of feature extraction, and provides the development process of 3D shapes feature extraction. Moreover, this paper introduces some common test datasets of 3D shapes, mainly discusses feature extraction algorithms of non-rigid 3D shapes, and prospects the future of 3D shapes feature extraction and feature matching.

Key words: feature extraction, non-rigid 3D shapes, curve-skeleton

摘要： 三维形状模型广泛应用于工业设计、教育、生物医药、动画娱乐、文物保护等多个领域中。三维形状模型的特征提取是计算机图形学和模式识别领域的重要问题，近年来受到学者的广泛关注。尤其是具有铰链、关节等的非刚性三维形状通常会发生变形，进一步增加了形状特征提取的难度。主要研究、分析、总结了近几年出现的刚性三维形状和非刚性三维形状的特征提取算法，分析了三维形状特征提取的难点，给出了三维形状特征提取的发展进程。介绍了近年来三维形状特征匹配研究领域中常用的一些测试数据库，重点探讨了非刚性三维形状的特征匹配方法，并展望了三维形状特征提取、特征匹配的未来发展方向。

关键词: 特征提取, 非刚性三维形状, 骨架