TOCSAC: TOpology Constraint SAmple Consensus for fast and reliable feature correspondence

Zhoucan He; Qing Wang; Heng Yang

doi:10.1007/978-3-642-10520-3_58

TOCSAC: TOpology Constraint SAmple Consensus for fast and reliable feature correspondence

Zhoucan He, Qing Wang, Heng Yang

School of Computer Science

Northwestern Polytechnical University Xian

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

This paper aims at outliers screening for the feature correspondence in image matching. A novel robust matching method, called topology constraint sample consensus (TOCSAC), is proposed to speed up the matching process while keeping the matching accuracy. The TOCSAC method comprises of two parts, the first of which is the constraint of points order, which should be invariant to scale, rotation and view point change. The second one is a constraint of affine invariant vector, which is also used to validate in similar and affine transforms. Comparing to the classical algorithms, such as RANSAC (random sample consensus) and PROSAC (progressive sample consensus), the proposed TOCSAC can significantly reduce time cost and improve the performance for wide base-line image correspondence.

Original language	English
Title of host publication	Advances in Visual Computing - 5th International Symposium, ISVC 2009, Proceedings
Pages	608-618
Number of pages	11
Edition	PART 2
DOIs	https://doi.org/10.1007/978-3-642-10520-3_58
State	Published - 2009
Event	5th International Symposium on Advances in Visual Computing, ISVC 2009 - Las Vegas, NV, United States Duration: 30 Nov 2009 → 2 Dec 2009

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Number	PART 2
Volume	5876 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	5th International Symposium on Advances in Visual Computing, ISVC 2009
Country/Territory	United States
City	Las Vegas, NV
Period	30/11/09 → 2/12/09

Access to Document

10.1007/978-3-642-10520-3_58

Cite this

He, Z., Wang, Q., & Yang, H. (2009). TOCSAC: TOpology Constraint SAmple Consensus for fast and reliable feature correspondence. In Advances in Visual Computing - 5th International Symposium, ISVC 2009, Proceedings (PART 2 ed., pp. 608-618). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5876 LNCS, No. PART 2). https://doi.org/10.1007/978-3-642-10520-3_58

He, Zhoucan ; Wang, Qing ; Yang, Heng. / TOCSAC : TOpology Constraint SAmple Consensus for fast and reliable feature correspondence. Advances in Visual Computing - 5th International Symposium, ISVC 2009, Proceedings. PART 2. ed. 2009. pp. 608-618 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 2).

@inproceedings{eac2b511f6e64700b7b18a0c49e73059,

title = "TOCSAC: TOpology Constraint SAmple Consensus for fast and reliable feature correspondence",

abstract = "This paper aims at outliers screening for the feature correspondence in image matching. A novel robust matching method, called topology constraint sample consensus (TOCSAC), is proposed to speed up the matching process while keeping the matching accuracy. The TOCSAC method comprises of two parts, the first of which is the constraint of points order, which should be invariant to scale, rotation and view point change. The second one is a constraint of affine invariant vector, which is also used to validate in similar and affine transforms. Comparing to the classical algorithms, such as RANSAC (random sample consensus) and PROSAC (progressive sample consensus), the proposed TOCSAC can significantly reduce time cost and improve the performance for wide base-line image correspondence.",

author = "Zhoucan He and Qing Wang and Heng Yang",

year = "2009",

doi = "10.1007/978-3-642-10520-3_58",

language = "英语",

isbn = "364210519X",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

number = "PART 2",

pages = "608--618",

booktitle = "Advances in Visual Computing - 5th International Symposium, ISVC 2009, Proceedings",

edition = "PART 2",

note = "5th International Symposium on Advances in Visual Computing, ISVC 2009 ; Conference date: 30-11-2009 Through 02-12-2009",

}

He, Z, Wang, Q & Yang, H 2009, TOCSAC: TOpology Constraint SAmple Consensus for fast and reliable feature correspondence. in Advances in Visual Computing - 5th International Symposium, ISVC 2009, Proceedings. PART 2 edn, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), no. PART 2, vol. 5876 LNCS, pp. 608-618, 5th International Symposium on Advances in Visual Computing, ISVC 2009, Las Vegas, NV, United States, 30/11/09. https://doi.org/10.1007/978-3-642-10520-3_58

TOCSAC: TOpology Constraint SAmple Consensus for fast and reliable feature correspondence. / He, Zhoucan; Wang, Qing; Yang, Heng.
Advances in Visual Computing - 5th International Symposium, ISVC 2009, Proceedings. PART 2. ed. 2009. p. 608-618 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5876 LNCS, No. PART 2).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - TOCSAC

T2 - 5th International Symposium on Advances in Visual Computing, ISVC 2009

AU - He, Zhoucan

AU - Wang, Qing

AU - Yang, Heng

PY - 2009

Y1 - 2009

N2 - This paper aims at outliers screening for the feature correspondence in image matching. A novel robust matching method, called topology constraint sample consensus (TOCSAC), is proposed to speed up the matching process while keeping the matching accuracy. The TOCSAC method comprises of two parts, the first of which is the constraint of points order, which should be invariant to scale, rotation and view point change. The second one is a constraint of affine invariant vector, which is also used to validate in similar and affine transforms. Comparing to the classical algorithms, such as RANSAC (random sample consensus) and PROSAC (progressive sample consensus), the proposed TOCSAC can significantly reduce time cost and improve the performance for wide base-line image correspondence.

AB - This paper aims at outliers screening for the feature correspondence in image matching. A novel robust matching method, called topology constraint sample consensus (TOCSAC), is proposed to speed up the matching process while keeping the matching accuracy. The TOCSAC method comprises of two parts, the first of which is the constraint of points order, which should be invariant to scale, rotation and view point change. The second one is a constraint of affine invariant vector, which is also used to validate in similar and affine transforms. Comparing to the classical algorithms, such as RANSAC (random sample consensus) and PROSAC (progressive sample consensus), the proposed TOCSAC can significantly reduce time cost and improve the performance for wide base-line image correspondence.

UR - http://www.scopus.com/inward/record.url?scp=72449156195&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-10520-3_58

DO - 10.1007/978-3-642-10520-3_58

M3 - 会议稿件

AN - SCOPUS:72449156195

SN - 364210519X

SN - 9783642105197

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 608

EP - 618

BT - Advances in Visual Computing - 5th International Symposium, ISVC 2009, Proceedings

Y2 - 30 November 2009 through 2 December 2009

ER -

He Z, Wang Q, Yang H. TOCSAC: TOpology Constraint SAmple Consensus for fast and reliable feature correspondence. In Advances in Visual Computing - 5th International Symposium, ISVC 2009, Proceedings. PART 2 ed. 2009. p. 608-618. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 2). doi: 10.1007/978-3-642-10520-3_58

TOCSAC: TOpology Constraint SAmple Consensus for fast and reliable feature correspondence

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this