Super-Resolution of Large Volumes of Sentinel-2 Images with High Performance Distributed Deep Learning

Run Zhang; Gabriele Cavallaro; Jenia Jitsev

doi:10.1109/IGARSS39084.2020.9323734

Super-Resolution of Large Volumes of Sentinel-2 Images with High Performance Distributed Deep Learning

Run Zhang, Gabriele Cavallaro, Jenia Jitsev

Faculty of Electrical and Computer Engineering

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

Abstract

This work proposes a novel distributed deep learning model for Remote Sensing (RS) images super-resolution. High Performance Computing (HPC) systems with GPUs are used to accelerate the learning of the unknown low to high resolution mapping from large volumes of Sentinel-2 data. The proposed deep learning model is based on self-attention mechanism and residual learning. The results demonstrate that state-of-the-art performance can be achieved by keeping the size of the model relatively small. Synchronous data parallelism is applied to scale up the training process without severe performance loss. Distributed training is thus shown to speed up learning substantially while keeping performance intact.

Original language	English
Title of host publication	2020 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2020 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	617-620
Number of pages	4
ISBN (Electronic)	9781728163741
DOIs	https://doi.org/10.1109/IGARSS39084.2020.9323734
Publication status	Published - 26 Sept 2020
Event	2020 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2020 - Virtual, Waikoloa, United States Duration: 26 Sept 2020 → 2 Oct 2020

Publication series

Name	International Geoscience and Remote Sensing Symposium (IGARSS)

Conference

Conference	2020 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2020
Country/Territory	United States
City	Virtual, Waikoloa
Period	26/09/20 → 2/10/20

Bibliographical note

Publisher Copyright:
© 2020 IEEE.

Other keywords

distributed deep learning
high performance computing
Sentinel-2
super-resolution

Access to Document

10.1109/IGARSS39084.2020.9323734

Cite this

Zhang, R., Cavallaro, G., & Jitsev, J. (2020). Super-Resolution of Large Volumes of Sentinel-2 Images with High Performance Distributed Deep Learning. In 2020 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2020 - Proceedings (pp. 617-620). Article 9323734 (International Geoscience and Remote Sensing Symposium (IGARSS)). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IGARSS39084.2020.9323734

Zhang, Run ; Cavallaro, Gabriele ; Jitsev, Jenia. / Super-Resolution of Large Volumes of Sentinel-2 Images with High Performance Distributed Deep Learning. 2020 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2020 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2020. pp. 617-620 (International Geoscience and Remote Sensing Symposium (IGARSS)).

@inproceedings{408204517e464aa984b0f5ab257999b2,

title = "Super-Resolution of Large Volumes of Sentinel-2 Images with High Performance Distributed Deep Learning",

abstract = "This work proposes a novel distributed deep learning model for Remote Sensing (RS) images super-resolution. High Performance Computing (HPC) systems with GPUs are used to accelerate the learning of the unknown low to high resolution mapping from large volumes of Sentinel-2 data. The proposed deep learning model is based on self-attention mechanism and residual learning. The results demonstrate that state-of-the-art performance can be achieved by keeping the size of the model relatively small. Synchronous data parallelism is applied to scale up the training process without severe performance loss. Distributed training is thus shown to speed up learning substantially while keeping performance intact.",

keywords = "distributed deep learning, high performance computing, Sentinel-2, super-resolution",

author = "Run Zhang and Gabriele Cavallaro and Jenia Jitsev",

note = "Publisher Copyright: {\textcopyright} 2020 IEEE.; 2020 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2020 ; Conference date: 26-09-2020 Through 02-10-2020",

year = "2020",

month = sep,

day = "26",

doi = "10.1109/IGARSS39084.2020.9323734",

language = "English",

series = "International Geoscience and Remote Sensing Symposium (IGARSS)",

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

pages = "617--620",

booktitle = "2020 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2020 - Proceedings",

address = "United States",

}

Zhang, R, Cavallaro, G & Jitsev, J 2020, Super-Resolution of Large Volumes of Sentinel-2 Images with High Performance Distributed Deep Learning. in 2020 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2020 - Proceedings., 9323734, International Geoscience and Remote Sensing Symposium (IGARSS), Institute of Electrical and Electronics Engineers Inc., pp. 617-620, 2020 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2020, Virtual, Waikoloa, United States, 26/09/20. https://doi.org/10.1109/IGARSS39084.2020.9323734

Super-Resolution of Large Volumes of Sentinel-2 Images with High Performance Distributed Deep Learning. / Zhang, Run; Cavallaro, Gabriele; Jitsev, Jenia.
2020 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2020 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2020. p. 617-620 9323734 (International Geoscience and Remote Sensing Symposium (IGARSS)).

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

TY - GEN

T1 - Super-Resolution of Large Volumes of Sentinel-2 Images with High Performance Distributed Deep Learning

AU - Zhang, Run

AU - Cavallaro, Gabriele

AU - Jitsev, Jenia

PY - 2020/9/26

Y1 - 2020/9/26

N2 - This work proposes a novel distributed deep learning model for Remote Sensing (RS) images super-resolution. High Performance Computing (HPC) systems with GPUs are used to accelerate the learning of the unknown low to high resolution mapping from large volumes of Sentinel-2 data. The proposed deep learning model is based on self-attention mechanism and residual learning. The results demonstrate that state-of-the-art performance can be achieved by keeping the size of the model relatively small. Synchronous data parallelism is applied to scale up the training process without severe performance loss. Distributed training is thus shown to speed up learning substantially while keeping performance intact.

AB - This work proposes a novel distributed deep learning model for Remote Sensing (RS) images super-resolution. High Performance Computing (HPC) systems with GPUs are used to accelerate the learning of the unknown low to high resolution mapping from large volumes of Sentinel-2 data. The proposed deep learning model is based on self-attention mechanism and residual learning. The results demonstrate that state-of-the-art performance can be achieved by keeping the size of the model relatively small. Synchronous data parallelism is applied to scale up the training process without severe performance loss. Distributed training is thus shown to speed up learning substantially while keeping performance intact.

KW - distributed deep learning

KW - high performance computing

KW - Sentinel-2

KW - super-resolution

UR - http://www.scopus.com/inward/record.url?scp=85101993700&partnerID=8YFLogxK

U2 - 10.1109/IGARSS39084.2020.9323734

DO - 10.1109/IGARSS39084.2020.9323734

M3 - Conference contribution

AN - SCOPUS:85101993700

T3 - International Geoscience and Remote Sensing Symposium (IGARSS)

SP - 617

EP - 620

BT - 2020 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2020 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2020 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2020

Y2 - 26 September 2020 through 2 October 2020

ER -

Zhang R, Cavallaro G, Jitsev J. Super-Resolution of Large Volumes of Sentinel-2 Images with High Performance Distributed Deep Learning. In 2020 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2020 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2020. p. 617-620. 9323734. (International Geoscience and Remote Sensing Symposium (IGARSS)). doi: 10.1109/IGARSS39084.2020.9323734

Super-Resolution of Large Volumes of Sentinel-2 Images with High Performance Distributed Deep Learning

Abstract

Publication series

Conference

Bibliographical note

Other keywords

Access to Document

Fingerprint

Cite this