An eigenvector approach based on shape context patterns for point matching

Xiabi Liu; Yunde Jia; Yanjie Wang

doi:10.1109/ISCIT.2006.339987

An eigenvector approach based on shape context patterns for point matching

Xiabi Liu^*
, Yunde Jia
, Yanjie Wang

^*Corresponding author for this work

School of Computer Science and Technology

Beijing Institute of Technology

Research output: Contribution to conference › Paper › peer-review

3 Citations (Scopus)

Abstract

In this paper, the problem of point correspondence across two images is treated in the eigenvector analysis matching framework of Scott and Longuet-Higgins. We develop the concept of shape contexts introduced by S. Belongie et al. to shape context patterns as rich local descriptors of points. We further propose a Gaussian-weighted Hausdorff distance between shape context patterns to measure correspondence strength in Scott and Longuet-Higgins framework. The resultant point matching approach is applied to estimate affine transformation between handwritten Chinese character images, whose effectiveness is confirmed by the experimental results.

Original language	English
Pages	455-458
Number of pages	4
DOIs	https://doi.org/10.1109/ISCIT.2006.339987
Publication status	Published - 2006
Event	2006 International Symposium on Communications and Information Technologies, ISCIT - Bangkok, Thailand Duration: 18 Oct 2006 → 20 Oct 2006

Conference

Conference	2006 International Symposium on Communications and Information Technologies, ISCIT
Country/Territory	Thailand
City	Bangkok
Period	18/10/06 → 20/10/06

Access to Document

10.1109/ISCIT.2006.339987

Cite this

@conference{5b767ad5a29744d4b48eb3aa3f0c451a,

title = "An eigenvector approach based on shape context patterns for point matching",

abstract = "In this paper, the problem of point correspondence across two images is treated in the eigenvector analysis matching framework of Scott and Longuet-Higgins. We develop the concept of shape contexts introduced by S. Belongie et al. to shape context patterns as rich local descriptors of points. We further propose a Gaussian-weighted Hausdorff distance between shape context patterns to measure correspondence strength in Scott and Longuet-Higgins framework. The resultant point matching approach is applied to estimate affine transformation between handwritten Chinese character images, whose effectiveness is confirmed by the experimental results.",

author = "Xiabi Liu and Yunde Jia and Yanjie Wang",

year = "2006",

doi = "10.1109/ISCIT.2006.339987",

language = "English",

pages = "455--458",

note = "2006 International Symposium on Communications and Information Technologies, ISCIT ; Conference date: 18-10-2006 Through 20-10-2006",

}

TY - CONF

T1 - An eigenvector approach based on shape context patterns for point matching

AU - Liu, Xiabi

AU - Jia, Yunde

AU - Wang, Yanjie

PY - 2006

Y1 - 2006

N2 - In this paper, the problem of point correspondence across two images is treated in the eigenvector analysis matching framework of Scott and Longuet-Higgins. We develop the concept of shape contexts introduced by S. Belongie et al. to shape context patterns as rich local descriptors of points. We further propose a Gaussian-weighted Hausdorff distance between shape context patterns to measure correspondence strength in Scott and Longuet-Higgins framework. The resultant point matching approach is applied to estimate affine transformation between handwritten Chinese character images, whose effectiveness is confirmed by the experimental results.

AB - In this paper, the problem of point correspondence across two images is treated in the eigenvector analysis matching framework of Scott and Longuet-Higgins. We develop the concept of shape contexts introduced by S. Belongie et al. to shape context patterns as rich local descriptors of points. We further propose a Gaussian-weighted Hausdorff distance between shape context patterns to measure correspondence strength in Scott and Longuet-Higgins framework. The resultant point matching approach is applied to estimate affine transformation between handwritten Chinese character images, whose effectiveness is confirmed by the experimental results.

UR - https://www.scopus.com/pages/publications/50949133826

U2 - 10.1109/ISCIT.2006.339987

DO - 10.1109/ISCIT.2006.339987

M3 - Paper

AN - SCOPUS:50949133826

SP - 455

EP - 458

T2 - 2006 International Symposium on Communications and Information Technologies, ISCIT

Y2 - 18 October 2006 through 20 October 2006

ER -

An eigenvector approach based on shape context patterns for point matching

Abstract

Conference

Access to Document

Other files and links

Fingerprint

Cite this