A note on the convergence of the mean shift

被引：49

作者：

Li, Xiangru ^{[1
]}

Hu, Zhanyi ^{[1
]}

Wu, Fuchao ^{[1
]}

机构：

[1] Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100080, Peoples R China

来源：

PATTERN RECOGNITION | 2007年 / 40卷 / 06期

基金：

中国国家自然科学基金;

关键词：

mean shift; convergence; local structure; computer vision;

D O I：

10.1016/j.patcog.2006.10.016

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Mean shift is an effective iterative algorithm widely used in computer vision community. However, to our knowledge, its convergence, a key aspect of any iterative algorithm, has not been rigorously proved up to now. In this paper, by further imposing some commonly acceptable conditions, its convergence is proved. (c) 2006 Published by Elsevier Ltd on behalf of Pattern Recognition Society.

引用

页码：1756 / 1762

页数：7

共 9 条

[1] MEAN SHIFT, MODE SEEKING, AND CLUSTERING [J].

CHENG, YZ .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1995, 17 (08) :790-799

[2] Kernel-based object tracking [J].

Comaniciu, D ;

Ramesh, V ;

Meer, P .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2003, 25 (05) :564-577

[3] Mean shift: A robust approach toward feature space analysis [J].

Comaniciu, D ;

Meer, P .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2002, 24 (05) :603-619

[4]

Comaniciu D, 2001, EIGHTH IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION, VOL I, PROCEEDINGS, P438, DOI 10.1109/ICCV.2001.937550

[5]

Comaniciu D, 2000, PROC CVPR IEEE, P142, DOI 10.1109/CVPR.2000.854761

[6] Mean shift is a bound optimization [J].

Fashing, M ;

Tomasi, C .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2005, 27 (03) :471-474

[7]

FUKUNAGA K, 1975, IEEE T INFORM THEORY, V21, P32, DOI 10.1109/TIT.1975.1055330

[8]

SINGH M, 2004, EUR C COMP VIS, V1, P508

[9]

XUE CX, 1993, REAL FUNCTION THEORY

← 1 →