Deep multi-spectral registration using invariant descriptor learning

Nati Ofir; Shai Silberstein; Hila Levi; Dani Rozenbaum; Yosi Keller; Sharon Duvdevani Bar

doi:10.1109/ICIP.2018.8451640

Deep multi-spectral registration using invariant descriptor learning

Nati Ofir, Shai Silberstein, Hila Levi, Dani Rozenbaum, Yosi Keller, Sharon Duvdevani Bar

Bar-Ilan University

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

6 Scopus citations

Abstract

In this work, we propose a deep-learning approach for aligning cross-spectral images. Our approach utilizes a learned descriptor invariant to different spectra. Multi-modal images of the same scene capture different characteristics and therefore their registration is challenging. To that end, we developed a feature-based approach for registering visible (VIS) to Near-Infra-Red (NIR) images. Our scheme detects corners by Harris and matches them by a patch-metric learned on top of a network trained using the CIFAR-10 dataset. As our experiments demonstrate, we achieve accurate alignment of cross-spectral images with sub-pixel accuracy. Comparing to contemporary state-of-the-art, our approach is more accurate in the task of VIS to NIR registration.

Original language	English
Title of host publication	2018 IEEE International Conference on Image Processing, ICIP 2018 - Proceedings
Publisher	IEEE Computer Society
Pages	1238-1242
Number of pages	5
ISBN (Electronic)	9781479970612
DOIs	https://doi.org/10.1109/ICIP.2018.8451640
State	Published - 29 Aug 2018
Event	25th IEEE International Conference on Image Processing, ICIP 2018 - Athens, Greece Duration: 7 Oct 2018 → 10 Oct 2018

Publication series

Name	Proceedings - International Conference on Image Processing, ICIP
ISSN (Print)	1522-4880

Conference

Conference	25th IEEE International Conference on Image Processing, ICIP 2018
Country/Territory	Greece
City	Athens
Period	7/10/18 → 10/10/18

Bibliographical note

Publisher Copyright:
© 2018 IEEE.

Keywords

Deep-Learning
Image Registration
Multi-Spectral Imaging

Access to Document

10.1109/ICIP.2018.8451640

Cite this

Ofir, N., Silberstein, S., Levi, H., Rozenbaum, D., Keller, Y., & Duvdevani Bar, S. (2018). Deep multi-spectral registration using invariant descriptor learning. In 2018 IEEE International Conference on Image Processing, ICIP 2018 - Proceedings (pp. 1238-1242). Article 8451640 (Proceedings - International Conference on Image Processing, ICIP). IEEE Computer Society. https://doi.org/10.1109/ICIP.2018.8451640

@inproceedings{b0ac538b261144c883c4c9839997af52,

title = "Deep multi-spectral registration using invariant descriptor learning",

abstract = "In this work, we propose a deep-learning approach for aligning cross-spectral images. Our approach utilizes a learned descriptor invariant to different spectra. Multi-modal images of the same scene capture different characteristics and therefore their registration is challenging. To that end, we developed a feature-based approach for registering visible (VIS) to Near-Infra-Red (NIR) images. Our scheme detects corners by Harris and matches them by a patch-metric learned on top of a network trained using the CIFAR-10 dataset. As our experiments demonstrate, we achieve accurate alignment of cross-spectral images with sub-pixel accuracy. Comparing to contemporary state-of-the-art, our approach is more accurate in the task of VIS to NIR registration.",

keywords = "Deep-Learning, Image Registration, Multi-Spectral Imaging",

author = "Nati Ofir and Shai Silberstein and Hila Levi and Dani Rozenbaum and Yosi Keller and {Duvdevani Bar}, Sharon",

note = "Publisher Copyright: {\textcopyright} 2018 IEEE.; 25th IEEE International Conference on Image Processing, ICIP 2018 ; Conference date: 07-10-2018 Through 10-10-2018",

year = "2018",

month = aug,

day = "29",

doi = "10.1109/ICIP.2018.8451640",

language = "אנגלית",

series = "Proceedings - International Conference on Image Processing, ICIP",

publisher = "IEEE Computer Society",

pages = "1238--1242",

booktitle = "2018 IEEE International Conference on Image Processing, ICIP 2018 - Proceedings",

address = "ארצות הברית",

}

Ofir, N, Silberstein, S, Levi, H, Rozenbaum, D, Keller, Y & Duvdevani Bar, S 2018, Deep multi-spectral registration using invariant descriptor learning. in 2018 IEEE International Conference on Image Processing, ICIP 2018 - Proceedings., 8451640, Proceedings - International Conference on Image Processing, ICIP, IEEE Computer Society, pp. 1238-1242, 25th IEEE International Conference on Image Processing, ICIP 2018, Athens, Greece, 7/10/18. https://doi.org/10.1109/ICIP.2018.8451640

Deep multi-spectral registration using invariant descriptor learning. / Ofir, Nati; Silberstein, Shai; Levi, Hila et al.
2018 IEEE International Conference on Image Processing, ICIP 2018 - Proceedings. IEEE Computer Society, 2018. p. 1238-1242 8451640 (Proceedings - International Conference on Image Processing, ICIP).

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

TY - GEN

T1 - Deep multi-spectral registration using invariant descriptor learning

AU - Ofir, Nati

AU - Silberstein, Shai

AU - Levi, Hila

AU - Rozenbaum, Dani

AU - Keller, Yosi

AU - Duvdevani Bar, Sharon

PY - 2018/8/29

Y1 - 2018/8/29

N2 - In this work, we propose a deep-learning approach for aligning cross-spectral images. Our approach utilizes a learned descriptor invariant to different spectra. Multi-modal images of the same scene capture different characteristics and therefore their registration is challenging. To that end, we developed a feature-based approach for registering visible (VIS) to Near-Infra-Red (NIR) images. Our scheme detects corners by Harris and matches them by a patch-metric learned on top of a network trained using the CIFAR-10 dataset. As our experiments demonstrate, we achieve accurate alignment of cross-spectral images with sub-pixel accuracy. Comparing to contemporary state-of-the-art, our approach is more accurate in the task of VIS to NIR registration.

AB - In this work, we propose a deep-learning approach for aligning cross-spectral images. Our approach utilizes a learned descriptor invariant to different spectra. Multi-modal images of the same scene capture different characteristics and therefore their registration is challenging. To that end, we developed a feature-based approach for registering visible (VIS) to Near-Infra-Red (NIR) images. Our scheme detects corners by Harris and matches them by a patch-metric learned on top of a network trained using the CIFAR-10 dataset. As our experiments demonstrate, we achieve accurate alignment of cross-spectral images with sub-pixel accuracy. Comparing to contemporary state-of-the-art, our approach is more accurate in the task of VIS to NIR registration.

KW - Deep-Learning

KW - Image Registration

KW - Multi-Spectral Imaging

UR - http://www.scopus.com/inward/record.url?scp=85062908138&partnerID=8YFLogxK

U2 - 10.1109/ICIP.2018.8451640

DO - 10.1109/ICIP.2018.8451640

M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???

AN - SCOPUS:85062908138

T3 - Proceedings - International Conference on Image Processing, ICIP

SP - 1238

EP - 1242

BT - 2018 IEEE International Conference on Image Processing, ICIP 2018 - Proceedings

PB - IEEE Computer Society

T2 - 25th IEEE International Conference on Image Processing, ICIP 2018

Y2 - 7 October 2018 through 10 October 2018

ER -

Ofir N, Silberstein S, Levi H, Rozenbaum D, Keller Y, Duvdevani Bar S. Deep multi-spectral registration using invariant descriptor learning. In 2018 IEEE International Conference on Image Processing, ICIP 2018 - Proceedings. IEEE Computer Society. 2018. p. 1238-1242. 8451640. (Proceedings - International Conference on Image Processing, ICIP). doi: 10.1109/ICIP.2018.8451640

Deep multi-spectral registration using invariant descriptor learning

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this