Home > Research > Publications & Outputs > Multi-stage Attention ResU-Net for Semantic Seg...

Electronic data

  • GRSL_01748_2020_final

    Rights statement: ©2021 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.

    Accepted author manuscript, 1.46 MB, PDF document

    Available under license: CC BY-NC: Creative Commons Attribution-NonCommercial 4.0 International License

Links

Text available via DOI:

View graph of relations

Multi-stage Attention ResU-Net for Semantic Segmentation of Fine-Resolution Remote Sensing Images

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published

Standard

Multi-stage Attention ResU-Net for Semantic Segmentation of Fine-Resolution Remote Sensing Images. / Li, Rui; Zheng, Shunyi; Duan, Chenxi et al.
In: IEEE Geoscience and Remote Sensing Letters, Vol. 19, 8009205, 01.01.2022.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

Li, R, Zheng, S, Duan, C, Su, J & Zhang, C 2022, 'Multi-stage Attention ResU-Net for Semantic Segmentation of Fine-Resolution Remote Sensing Images', IEEE Geoscience and Remote Sensing Letters, vol. 19, 8009205. https://doi.org/10.1109/LGRS.2021.3063381

APA

Li, R., Zheng, S., Duan, C., Su, J., & Zhang, C. (2022). Multi-stage Attention ResU-Net for Semantic Segmentation of Fine-Resolution Remote Sensing Images. IEEE Geoscience and Remote Sensing Letters, 19, Article 8009205. https://doi.org/10.1109/LGRS.2021.3063381

Vancouver

Li R, Zheng S, Duan C, Su J, Zhang C. Multi-stage Attention ResU-Net for Semantic Segmentation of Fine-Resolution Remote Sensing Images. IEEE Geoscience and Remote Sensing Letters. 2022 Jan 1;19:8009205. Epub 2021 Mar 15. doi: 10.1109/LGRS.2021.3063381

Author

Li, Rui ; Zheng, Shunyi ; Duan, Chenxi et al. / Multi-stage Attention ResU-Net for Semantic Segmentation of Fine-Resolution Remote Sensing Images. In: IEEE Geoscience and Remote Sensing Letters. 2022 ; Vol. 19.

Bibtex

@article{930a2adfdfec4e78b11a7eeff9c5bfe7,
title = "Multi-stage Attention ResU-Net for Semantic Segmentation of Fine-Resolution Remote Sensing Images",
abstract = "The attention mechanism can refine the extracted feature maps and boost the classification performance of the deep network, which has become an essential technique in computer vision and natural language processing. However, the memory and computational costs of the dot-product attention mechanism increase quadratically with the spatio-temporal size of the input. Such growth hinders the usage of attention mechanisms considerably in application scenarios with large-scale inputs. In this Letter, we propose a Linear Attention Mechanism (LAM) to address this issue, which is approximately equivalent to dot-product attention with computational efficiency. Such a design makes the incorporation between attention mechanisms and deep networks much more flexible and versatile. Based on the proposed LAM, we re-factor the skip connections in the raw U-Net and design a Multi-stage Attention ResU-Net (MAResU-Net) for semantic segmentation from fine-resolution remote sensing images. Experiments conducted on the Vaihingen dataset demonstrated the effectiveness and efficiency of our MAResU-Net. Our code is available at https://github.com/lironui/MAResU-Net.",
keywords = "semantic segmentation, fine-resolution remote sensing images, linear attention mechanism",
author = "Rui Li and Shunyi Zheng and Chenxi Duan and Jianlin Su and Ce Zhang",
note = "{\textcopyright}2021 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE. ",
year = "2022",
month = jan,
day = "1",
doi = "10.1109/LGRS.2021.3063381",
language = "English",
volume = "19",
journal = "IEEE Geoscience and Remote Sensing Letters",
issn = "1545-598X",
publisher = "IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC",

}

RIS

TY - JOUR

T1 - Multi-stage Attention ResU-Net for Semantic Segmentation of Fine-Resolution Remote Sensing Images

AU - Li, Rui

AU - Zheng, Shunyi

AU - Duan, Chenxi

AU - Su, Jianlin

AU - Zhang, Ce

N1 - ©2021 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.

PY - 2022/1/1

Y1 - 2022/1/1

N2 - The attention mechanism can refine the extracted feature maps and boost the classification performance of the deep network, which has become an essential technique in computer vision and natural language processing. However, the memory and computational costs of the dot-product attention mechanism increase quadratically with the spatio-temporal size of the input. Such growth hinders the usage of attention mechanisms considerably in application scenarios with large-scale inputs. In this Letter, we propose a Linear Attention Mechanism (LAM) to address this issue, which is approximately equivalent to dot-product attention with computational efficiency. Such a design makes the incorporation between attention mechanisms and deep networks much more flexible and versatile. Based on the proposed LAM, we re-factor the skip connections in the raw U-Net and design a Multi-stage Attention ResU-Net (MAResU-Net) for semantic segmentation from fine-resolution remote sensing images. Experiments conducted on the Vaihingen dataset demonstrated the effectiveness and efficiency of our MAResU-Net. Our code is available at https://github.com/lironui/MAResU-Net.

AB - The attention mechanism can refine the extracted feature maps and boost the classification performance of the deep network, which has become an essential technique in computer vision and natural language processing. However, the memory and computational costs of the dot-product attention mechanism increase quadratically with the spatio-temporal size of the input. Such growth hinders the usage of attention mechanisms considerably in application scenarios with large-scale inputs. In this Letter, we propose a Linear Attention Mechanism (LAM) to address this issue, which is approximately equivalent to dot-product attention with computational efficiency. Such a design makes the incorporation between attention mechanisms and deep networks much more flexible and versatile. Based on the proposed LAM, we re-factor the skip connections in the raw U-Net and design a Multi-stage Attention ResU-Net (MAResU-Net) for semantic segmentation from fine-resolution remote sensing images. Experiments conducted on the Vaihingen dataset demonstrated the effectiveness and efficiency of our MAResU-Net. Our code is available at https://github.com/lironui/MAResU-Net.

KW - semantic segmentation

KW - fine-resolution remote sensing images

KW - linear attention mechanism

U2 - 10.1109/LGRS.2021.3063381

DO - 10.1109/LGRS.2021.3063381

M3 - Journal article

VL - 19

JO - IEEE Geoscience and Remote Sensing Letters

JF - IEEE Geoscience and Remote Sensing Letters

SN - 1545-598X

M1 - 8009205

ER -