Disentangling the Benefits of Self-Supervised Learning to Deployment-Driven Downstream Tasks of Satellite Images

Zhuo Deng; Yibing Wei; Mingye Zhu; Xueliang Wang; Junchi Zhou; Zhicheng Yang; Hang Zhou; Zhenjie Cao; Lan Ma; Mei Han; Jui Hsin Lai

Disentangling the Benefits of Self-Supervised Learning to Deployment-Driven Downstream Tasks of Satellite Images

Zhuo Deng, Yibing Wei, Mingye Zhu, Xueliang Wang, Junchi Zhou, Zhicheng Yang, Hang Zhou, Zhenjie Cao, Lan Ma, Mei Han, Jui Hsin Lai

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

Abstract

In this paper, we investigate the benefits of self-supervised learning (SSL) to downstream tasks of satellite images. Unlike common student academic projects, this work focuses on the advantages of the SSL for deployment-driven tasks which have specific scenarios with low or high-spatial resolution images. Our preliminary experiments demonstrate the robust benefits of the SSL trained by medium-resolution (10m) images to both low-resolution (100m) scene classification case (4.25%↑) and very high-resolution (5cm) aerial image segmentation case (1.96%↑), respectively.

Original language	English (US)
Title of host publication	AAAI-23 Special Programs, IAAI-23, EAAI-23, Student Papers and Demonstrations
Editors	Brian Williams, Yiling Chen, Jennifer Neville
Publisher	AAAI Press
Pages	16198-16199
Number of pages	2
ISBN (Electronic)	9781577358800
State	Published - Jun 27 2023
Event	37th AAAI Conference on Artificial Intelligence, AAAI 2023 - Washington, United States Duration: Feb 7 2023 → Feb 14 2023

Publication series

Name	Proceedings of the 37th AAAI Conference on Artificial Intelligence, AAAI 2023
Volume	37

Conference

Conference	37th AAAI Conference on Artificial Intelligence, AAAI 2023
Country/Territory	United States
City	Washington
Period	2/7/23 → 2/14/23

ASJC Scopus subject areas

Artificial Intelligence

Cite this

Deng, Z., Wei, Y., Zhu, M., Wang, X., Zhou, J., Yang, Z., Zhou, H., Cao, Z., Ma, L., Han, M., & Lai, J. H. (2023). Disentangling the Benefits of Self-Supervised Learning to Deployment-Driven Downstream Tasks of Satellite Images. In B. Williams, Y. Chen, & J. Neville (Eds.), AAAI-23 Special Programs, IAAI-23, EAAI-23, Student Papers and Demonstrations (pp. 16198-16199). (Proceedings of the 37th AAAI Conference on Artificial Intelligence, AAAI 2023; Vol. 37). AAAI Press.

Disentangling the Benefits of Self-Supervised Learning to Deployment-Driven Downstream Tasks of Satellite Images. / Deng, Zhuo; Wei, Yibing; Zhu, Mingye et al.
AAAI-23 Special Programs, IAAI-23, EAAI-23, Student Papers and Demonstrations. ed. / Brian Williams; Yiling Chen; Jennifer Neville. AAAI Press, 2023. p. 16198-16199 (Proceedings of the 37th AAAI Conference on Artificial Intelligence, AAAI 2023; Vol. 37).

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

Deng, Z, Wei, Y, Zhu, M, Wang, X, Zhou, J, Yang, Z, Zhou, H, Cao, Z, Ma, L, Han, M & Lai, JH 2023, Disentangling the Benefits of Self-Supervised Learning to Deployment-Driven Downstream Tasks of Satellite Images. in B Williams, Y Chen & J Neville (eds), AAAI-23 Special Programs, IAAI-23, EAAI-23, Student Papers and Demonstrations. Proceedings of the 37th AAAI Conference on Artificial Intelligence, AAAI 2023, vol. 37, AAAI Press, pp. 16198-16199, 37th AAAI Conference on Artificial Intelligence, AAAI 2023, Washington, United States, 2/7/23.

Deng Z, Wei Y, Zhu M, Wang X, Zhou J, Yang Z et al. Disentangling the Benefits of Self-Supervised Learning to Deployment-Driven Downstream Tasks of Satellite Images. In Williams B, Chen Y, Neville J, editors, AAAI-23 Special Programs, IAAI-23, EAAI-23, Student Papers and Demonstrations. AAAI Press. 2023. p. 16198-16199. (Proceedings of the 37th AAAI Conference on Artificial Intelligence, AAAI 2023).

Deng, Zhuo ; Wei, Yibing ; Zhu, Mingye et al. / Disentangling the Benefits of Self-Supervised Learning to Deployment-Driven Downstream Tasks of Satellite Images. AAAI-23 Special Programs, IAAI-23, EAAI-23, Student Papers and Demonstrations. editor / Brian Williams ; Yiling Chen ; Jennifer Neville. AAAI Press, 2023. pp. 16198-16199 (Proceedings of the 37th AAAI Conference on Artificial Intelligence, AAAI 2023).

@inproceedings{9caf1b15ccb042a1b62a9a18ee123811,

title = "Disentangling the Benefits of Self-Supervised Learning to Deployment-Driven Downstream Tasks of Satellite Images",

abstract = "In this paper, we investigate the benefits of self-supervised learning (SSL) to downstream tasks of satellite images. Unlike common student academic projects, this work focuses on the advantages of the SSL for deployment-driven tasks which have specific scenarios with low or high-spatial resolution images. Our preliminary experiments demonstrate the robust benefits of the SSL trained by medium-resolution (10m) images to both low-resolution (100m) scene classification case (4.25%↑) and very high-resolution (5cm) aerial image segmentation case (1.96%↑), respectively.",

author = "Zhuo Deng and Yibing Wei and Mingye Zhu and Xueliang Wang and Junchi Zhou and Zhicheng Yang and Hang Zhou and Zhenjie Cao and Lan Ma and Mei Han and Lai, {Jui Hsin}",

note = "Publisher Copyright: Copyright {\textcopyright} 2023, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved.; 37th AAAI Conference on Artificial Intelligence, AAAI 2023 ; Conference date: 07-02-2023 Through 14-02-2023",

year = "2023",

month = jun,

day = "27",

language = "English (US)",

series = "Proceedings of the 37th AAAI Conference on Artificial Intelligence, AAAI 2023",

publisher = "AAAI Press",

pages = "16198--16199",

editor = "Brian Williams and Yiling Chen and Jennifer Neville",

booktitle = "AAAI-23 Special Programs, IAAI-23, EAAI-23, Student Papers and Demonstrations",

}

TY - GEN

T1 - Disentangling the Benefits of Self-Supervised Learning to Deployment-Driven Downstream Tasks of Satellite Images

AU - Deng, Zhuo

AU - Wei, Yibing

AU - Zhu, Mingye

AU - Wang, Xueliang

AU - Zhou, Junchi

AU - Yang, Zhicheng

AU - Zhou, Hang

AU - Cao, Zhenjie

AU - Ma, Lan

AU - Han, Mei

AU - Lai, Jui Hsin

PY - 2023/6/27

Y1 - 2023/6/27

N2 - In this paper, we investigate the benefits of self-supervised learning (SSL) to downstream tasks of satellite images. Unlike common student academic projects, this work focuses on the advantages of the SSL for deployment-driven tasks which have specific scenarios with low or high-spatial resolution images. Our preliminary experiments demonstrate the robust benefits of the SSL trained by medium-resolution (10m) images to both low-resolution (100m) scene classification case (4.25%↑) and very high-resolution (5cm) aerial image segmentation case (1.96%↑), respectively.

AB - In this paper, we investigate the benefits of self-supervised learning (SSL) to downstream tasks of satellite images. Unlike common student academic projects, this work focuses on the advantages of the SSL for deployment-driven tasks which have specific scenarios with low or high-spatial resolution images. Our preliminary experiments demonstrate the robust benefits of the SSL trained by medium-resolution (10m) images to both low-resolution (100m) scene classification case (4.25%↑) and very high-resolution (5cm) aerial image segmentation case (1.96%↑), respectively.

UR - http://www.scopus.com/inward/record.url?scp=85168255306&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85168255306&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:85168255306

T3 - Proceedings of the 37th AAAI Conference on Artificial Intelligence, AAAI 2023

SP - 16198

EP - 16199

BT - AAAI-23 Special Programs, IAAI-23, EAAI-23, Student Papers and Demonstrations

A2 - Williams, Brian

A2 - Chen, Yiling

A2 - Neville, Jennifer

PB - AAAI Press

T2 - 37th AAAI Conference on Artificial Intelligence, AAAI 2023

Y2 - 7 February 2023 through 14 February 2023

ER -

Disentangling the Benefits of Self-Supervised Learning to Deployment-Driven Downstream Tasks of Satellite Images

Abstract

Publication series

Conference

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this