Part-level object recognition using superquadrics

被引：27

作者：

Krivic, J ^{[1
]}

Solina, F ^{[1
]}

机构：

[1] Univ Ljubljana, Fac Comp & Informat Sci, Comp Vis Lab, Ljubljana 1000, Slovenia

来源：

COMPUTER VISION AND IMAGE UNDERSTANDING | 2004年 / 95卷 / 01期

关键词：

superquadrics; part-level object modelling; range images; object recognition;

D O I：

10.1016/j.cviu.2003.11.002

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

This paper proposes a technique for object recognition using superquadric built models. Superquadrics, which are three-dimensional models suitable for part-level representation of objects, are reconstructed from range images using the recover-and-select paradigm. Using interpretation trees, the presence of an object from the model database can be hypothesized. These hypotheses are verified by projecting and re-fitting the object model to the range image of the scene which at the same time enables a better localization of the object in the scene. (C) 2004 Elsevier Inc. All rights reserved.

引用

页码：105 / 126

页数：22