ABFT

Anisotropic binary feature transform based on structure tensor space

Seungryong Kim, Hunjae Yoo, Seungchul Ryu, Bumsub Ham, Kwanghoon Sohn

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

Local feature matching is a fundamental step for many computer vision applications. Recently, binary feature transforms have been popularly proposed to improve the computational efficiency while preserving high matching performance. However, it is sensitive to noise and geometrical distortion such as affine transformation. In this paper, we propose ABFT framework, composed of a noise robust feature detection and affine invariant binary feature description based on a structure tensor space. Experimental results show that ABFT outperforms other state-of-the-art feature transforms in terms of the repeatability, recognition rate, and computational time.

Original languageEnglish
Title of host publication2013 IEEE International Conference on Image Processing, ICIP 2013 - Proceedings
Pages2920-2923
Number of pages4
DOIs
Publication statusPublished - 2013 Dec 1
Event2013 20th IEEE International Conference on Image Processing, ICIP 2013 - Melbourne, VIC, Australia
Duration: 2013 Sep 152013 Sep 18

Other

Other2013 20th IEEE International Conference on Image Processing, ICIP 2013
CountryAustralia
CityMelbourne, VIC
Period13/9/1513/9/18

Fingerprint

Computational efficiency
Computer vision
Tensors

All Science Journal Classification (ASJC) codes

  • Computer Vision and Pattern Recognition

Cite this

Kim, S., Yoo, H., Ryu, S., Ham, B., & Sohn, K. (2013). ABFT: Anisotropic binary feature transform based on structure tensor space. In 2013 IEEE International Conference on Image Processing, ICIP 2013 - Proceedings (pp. 2920-2923). [6738601] https://doi.org/10.1109/ICIP.2013.6738601
Kim, Seungryong ; Yoo, Hunjae ; Ryu, Seungchul ; Ham, Bumsub ; Sohn, Kwanghoon. / ABFT : Anisotropic binary feature transform based on structure tensor space. 2013 IEEE International Conference on Image Processing, ICIP 2013 - Proceedings. 2013. pp. 2920-2923
@inproceedings{ab5e73ba036149dfb6284dfe26a94d1e,
title = "ABFT: Anisotropic binary feature transform based on structure tensor space",
abstract = "Local feature matching is a fundamental step for many computer vision applications. Recently, binary feature transforms have been popularly proposed to improve the computational efficiency while preserving high matching performance. However, it is sensitive to noise and geometrical distortion such as affine transformation. In this paper, we propose ABFT framework, composed of a noise robust feature detection and affine invariant binary feature description based on a structure tensor space. Experimental results show that ABFT outperforms other state-of-the-art feature transforms in terms of the repeatability, recognition rate, and computational time.",
author = "Seungryong Kim and Hunjae Yoo and Seungchul Ryu and Bumsub Ham and Kwanghoon Sohn",
year = "2013",
month = "12",
day = "1",
doi = "10.1109/ICIP.2013.6738601",
language = "English",
isbn = "9781479923410",
pages = "2920--2923",
booktitle = "2013 IEEE International Conference on Image Processing, ICIP 2013 - Proceedings",

}

Kim, S, Yoo, H, Ryu, S, Ham, B & Sohn, K 2013, ABFT: Anisotropic binary feature transform based on structure tensor space. in 2013 IEEE International Conference on Image Processing, ICIP 2013 - Proceedings., 6738601, pp. 2920-2923, 2013 20th IEEE International Conference on Image Processing, ICIP 2013, Melbourne, VIC, Australia, 13/9/15. https://doi.org/10.1109/ICIP.2013.6738601

ABFT : Anisotropic binary feature transform based on structure tensor space. / Kim, Seungryong; Yoo, Hunjae; Ryu, Seungchul; Ham, Bumsub; Sohn, Kwanghoon.

2013 IEEE International Conference on Image Processing, ICIP 2013 - Proceedings. 2013. p. 2920-2923 6738601.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - ABFT

T2 - Anisotropic binary feature transform based on structure tensor space

AU - Kim, Seungryong

AU - Yoo, Hunjae

AU - Ryu, Seungchul

AU - Ham, Bumsub

AU - Sohn, Kwanghoon

PY - 2013/12/1

Y1 - 2013/12/1

N2 - Local feature matching is a fundamental step for many computer vision applications. Recently, binary feature transforms have been popularly proposed to improve the computational efficiency while preserving high matching performance. However, it is sensitive to noise and geometrical distortion such as affine transformation. In this paper, we propose ABFT framework, composed of a noise robust feature detection and affine invariant binary feature description based on a structure tensor space. Experimental results show that ABFT outperforms other state-of-the-art feature transforms in terms of the repeatability, recognition rate, and computational time.

AB - Local feature matching is a fundamental step for many computer vision applications. Recently, binary feature transforms have been popularly proposed to improve the computational efficiency while preserving high matching performance. However, it is sensitive to noise and geometrical distortion such as affine transformation. In this paper, we propose ABFT framework, composed of a noise robust feature detection and affine invariant binary feature description based on a structure tensor space. Experimental results show that ABFT outperforms other state-of-the-art feature transforms in terms of the repeatability, recognition rate, and computational time.

UR - http://www.scopus.com/inward/record.url?scp=84897808832&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84897808832&partnerID=8YFLogxK

U2 - 10.1109/ICIP.2013.6738601

DO - 10.1109/ICIP.2013.6738601

M3 - Conference contribution

SN - 9781479923410

SP - 2920

EP - 2923

BT - 2013 IEEE International Conference on Image Processing, ICIP 2013 - Proceedings

ER -

Kim S, Yoo H, Ryu S, Ham B, Sohn K. ABFT: Anisotropic binary feature transform based on structure tensor space. In 2013 IEEE International Conference on Image Processing, ICIP 2013 - Proceedings. 2013. p. 2920-2923. 6738601 https://doi.org/10.1109/ICIP.2013.6738601