Integration of Shape Context and Semigroup Kernel in image classification

Yi Guo; Junbin Gao

doi:10.1109/ICMLC.2007.4370137

Integration of Shape Context and Semigroup Kernel in image classification

Yi Guo, Junbin Gao

Computing and Mathematics

Research output: Book chapter/Published conference paper › Conference paper › peer-review

4 Citations (Scopus)

8 Downloads (Pure)

Abstract

Shape context is a rich descriptor for shapes and can be exploited to find pointwise correspondences between shapes, and thereby to obtain shape alignment by Thin Plate Spline (TPS). It is invariant under scaling and translation and robust under small geometrical distortions and presence of outliers. These features will supply a gap of the defect of semigroup kernel for its weakness in dealing with the deformation of the image. This paper integrates these two methods by defining a new kernel on shapes and images which is the combination of the shape distance from shape context and image similarity from semigroup kernel. Experiments of SVM classification on handwritten digits showed that it outperforms other existing kernels and the result of the data visualization exhibited another successful application of this new kernel.

Original language	English
Title of host publication	IEEE International Conference on Machine Learning and Cybernetics (ICMLC)
Editors	Xizhao Wang, Daniel Yeung
Place of Publication	Washington DC, USA
Publisher	IEEE
Pages	181-186
Number of pages	6
Volume	1
ISBN (Electronic)	142440973X
DOIs	https://doi.org/10.1109/ICMLC.2007.4370137
Publication status	Published - 2007
Event	ICMLC2007: 6th International Conference - Hotel Miramar, Hong Kong, China Duration: 19 Aug 2007 → 22 Aug 2007 http://www.hitsz.edu.cn/article/view/id-18804.html (conference info)

Conference

Conference	ICMLC2007: 6th International Conference
Country	China
City	Hong Kong
Period	19/08/07 → 22/08/07
Internet address	http://www.hitsz.edu.cn/article/view/id-18804.html (conference info)

Access to Document

10.1109/ICMLC.2007.4370137

CSU264097Accepted author manuscript, 2.5 MB

Fingerprint Dive into the research topics of 'Integration of Shape Context and Semigroup Kernel in image classification'. Together they form a unique fingerprint.

Cite this

@inproceedings{d07d0e7aa54c49be9e23c79131160947,

title = "Integration of Shape Context and Semigroup Kernel in image classification",

abstract = "Shape context is a rich descriptor for shapes and can be exploited to find pointwise correspondences between shapes, and thereby to obtain shape alignment by Thin Plate Spline (TPS). It is invariant under scaling and translation and robust under small geometrical distortions and presence of outliers. These features will supply a gap of the defect of semigroup kernel for its weakness in dealing with the deformation of the image. This paper integrates these two methods by defining a new kernel on shapes and images which is the combination of the shape distance from shape context and image similarity from semigroup kernel. Experiments of SVM classification on handwritten digits showed that it outperforms other existing kernels and the result of the data visualization exhibited another successful application of this new kernel.",

keywords = "Open access version available, Classification, Semigroup Kernel, Shape Context",

author = "Yi Guo and Junbin Gao",

note = "Imported on 03 May 2017 - DigiTool details were: publisher = Washington DC, USA: IEEE, 2007. editor/s (773b) = Wang, Xizhao; Yeung, Daniel; Event dates (773o) = 19-27 August, 2007; Parent title (773t) = IEEE International Conference on Machine Learning and Cybernetics (ICMLC).; ICMLC2007: 6th International Conference ; Conference date: 19-08-2007 Through 22-08-2007",

year = "2007",

doi = "10.1109/ICMLC.2007.4370137",

language = "English",

volume = "1",

pages = "181--186",

editor = "Xizhao Wang and Daniel Yeung",

booktitle = "IEEE International Conference on Machine Learning and Cybernetics (ICMLC)",

publisher = "IEEE",

url = "http://www.hitsz.edu.cn/article/view/id-18804.html",

}

TY - GEN

T1 - Integration of Shape Context and Semigroup Kernel in image classification

AU - Guo, Yi

AU - Gao, Junbin

N1 - Imported on 03 May 2017 - DigiTool details were: publisher = Washington DC, USA: IEEE, 2007. editor/s (773b) = Wang, Xizhao; Yeung, Daniel; Event dates (773o) = 19-27 August, 2007; Parent title (773t) = IEEE International Conference on Machine Learning and Cybernetics (ICMLC).

PY - 2007

Y1 - 2007

N2 - Shape context is a rich descriptor for shapes and can be exploited to find pointwise correspondences between shapes, and thereby to obtain shape alignment by Thin Plate Spline (TPS). It is invariant under scaling and translation and robust under small geometrical distortions and presence of outliers. These features will supply a gap of the defect of semigroup kernel for its weakness in dealing with the deformation of the image. This paper integrates these two methods by defining a new kernel on shapes and images which is the combination of the shape distance from shape context and image similarity from semigroup kernel. Experiments of SVM classification on handwritten digits showed that it outperforms other existing kernels and the result of the data visualization exhibited another successful application of this new kernel.

AB - Shape context is a rich descriptor for shapes and can be exploited to find pointwise correspondences between shapes, and thereby to obtain shape alignment by Thin Plate Spline (TPS). It is invariant under scaling and translation and robust under small geometrical distortions and presence of outliers. These features will supply a gap of the defect of semigroup kernel for its weakness in dealing with the deformation of the image. This paper integrates these two methods by defining a new kernel on shapes and images which is the combination of the shape distance from shape context and image similarity from semigroup kernel. Experiments of SVM classification on handwritten digits showed that it outperforms other existing kernels and the result of the data visualization exhibited another successful application of this new kernel.

KW - Open access version available

KW - Classification

KW - Semigroup Kernel

KW - Shape Context

U2 - 10.1109/ICMLC.2007.4370137

DO - 10.1109/ICMLC.2007.4370137

M3 - Conference paper

VL - 1

SP - 181

EP - 186

BT - IEEE International Conference on Machine Learning and Cybernetics (ICMLC)

A2 - Wang, Xizhao

A2 - Yeung, Daniel

PB - IEEE

CY - Washington DC, USA

T2 - ICMLC2007: 6th International Conference

Y2 - 19 August 2007 through 22 August 2007

ER -