Self-Supervised Learning in Remote Sensing: A review

Yi Wang; Conrad M. Albrecht; Nassim Ait Ali Braham; Lichao Mou; Xiao Xiang Zhu

doi:10.1109/MGRS.2022.3198244

Self-Supervised Learning in Remote Sensing: A review

Yi Wang, Conrad M. Albrecht, Nassim Ait Ali Braham, Lichao Mou, Xiao Xiang Zhu

Research output: Contribution to journal › Article › peer-review

154 Scopus citations

Abstract

In deep learning research, self-supervised learning (SSL) has received great attention, triggering interest within both the computer vision and remote sensing communities. While there has been big success in computer vision, most of the potential of SSL in the domain of Earth observation remains locked. In this article, we provide an introduction to and a review of the concepts and latest developments in SSL for computer vision in the context of remote sensing. Further, we provide a preliminary benchmark of modern SSL algorithms on popular remote sensing datasets, verifying the potential of SSL in remote sensing and providing an extended study on data augmentations. Finally, we identify a list of promising directions of future research in SSL for Earth observation (SSL4EO) to pave the way for the fruitful interaction of both domains.

Original language	English
Pages (from-to)	213-247
Number of pages	35
Journal	IEEE Geoscience and Remote Sensing Magazine
Volume	10
Issue number	4
DOIs	https://doi.org/10.1109/MGRS.2022.3198244
State	Published - 1 Dec 2022
Externally published	Yes

Access to Document

10.1109/MGRS.2022.3198244

Cite this

@article{dbf7981ea8af47d79b9c1d410b836a33,

title = "Self-Supervised Learning in Remote Sensing: A review",

abstract = "In deep learning research, self-supervised learning (SSL) has received great attention, triggering interest within both the computer vision and remote sensing communities. While there has been big success in computer vision, most of the potential of SSL in the domain of Earth observation remains locked. In this article, we provide an introduction to and a review of the concepts and latest developments in SSL for computer vision in the context of remote sensing. Further, we provide a preliminary benchmark of modern SSL algorithms on popular remote sensing datasets, verifying the potential of SSL in remote sensing and providing an extended study on data augmentations. Finally, we identify a list of promising directions of future research in SSL for Earth observation (SSL4EO) to pave the way for the fruitful interaction of both domains.",

author = "Yi Wang and Albrecht, {Conrad M.} and Braham, {Nassim Ait Ali} and Lichao Mou and Zhu, {Xiao Xiang}",

note = "Publisher Copyright: {\textcopyright} 2013 IEEE.",

year = "2022",

month = dec,

day = "1",

doi = "10.1109/MGRS.2022.3198244",

language = "English",

volume = "10",

pages = "213--247",

journal = "IEEE Geoscience and Remote Sensing Magazine",

issn = "2473-2397",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "4",

}

TY - JOUR

T1 - Self-Supervised Learning in Remote Sensing

T2 - A review

AU - Wang, Yi

AU - Albrecht, Conrad M.

AU - Braham, Nassim Ait Ali

AU - Mou, Lichao

AU - Zhu, Xiao Xiang

PY - 2022/12/1

Y1 - 2022/12/1

N2 - In deep learning research, self-supervised learning (SSL) has received great attention, triggering interest within both the computer vision and remote sensing communities. While there has been big success in computer vision, most of the potential of SSL in the domain of Earth observation remains locked. In this article, we provide an introduction to and a review of the concepts and latest developments in SSL for computer vision in the context of remote sensing. Further, we provide a preliminary benchmark of modern SSL algorithms on popular remote sensing datasets, verifying the potential of SSL in remote sensing and providing an extended study on data augmentations. Finally, we identify a list of promising directions of future research in SSL for Earth observation (SSL4EO) to pave the way for the fruitful interaction of both domains.

AB - In deep learning research, self-supervised learning (SSL) has received great attention, triggering interest within both the computer vision and remote sensing communities. While there has been big success in computer vision, most of the potential of SSL in the domain of Earth observation remains locked. In this article, we provide an introduction to and a review of the concepts and latest developments in SSL for computer vision in the context of remote sensing. Further, we provide a preliminary benchmark of modern SSL algorithms on popular remote sensing datasets, verifying the potential of SSL in remote sensing and providing an extended study on data augmentations. Finally, we identify a list of promising directions of future research in SSL for Earth observation (SSL4EO) to pave the way for the fruitful interaction of both domains.

UR - http://www.scopus.com/inward/record.url?scp=85137874425&partnerID=8YFLogxK

U2 - 10.1109/MGRS.2022.3198244

DO - 10.1109/MGRS.2022.3198244

M3 - Article

AN - SCOPUS:85137874425

SN - 2473-2397

VL - 10

SP - 213

EP - 247

JO - IEEE Geoscience and Remote Sensing Magazine

JF - IEEE Geoscience and Remote Sensing Magazine

IS - 4

ER -

Self-Supervised Learning in Remote Sensing: A review

Abstract

Access to Document

Other files and links

Fingerprint

Cite this