A TRANSFORMER-BASED SIAMESE NETWORK FOR CHANGE DETECTION

被引：513
作者：
Bandara, Wele Gedara Chaminda [1 ]
Patel, Vishal M. [1 ]
机构：
[1] Johns Hopkins Univ, Baltimore, MD USA
来源：
2022 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM (IGARSS 2022) | 2022年
关键词：
Change detection; transformer Siamese network; attention mechanism; multilayer perceptron; remote sensing;
D O I：
10.1109/IGARSS46834.2022.9883686
中图分类号：
P [天文学、地球科学];
学科分类号：
07 ;
摘要：
This paper presents a transformer-based Siamese network architecture (abbreviated by ChangeFormer) for Change Detection (CD) from a pair of co-registered remote sensing images. Different from recent CD frameworks, which are based on fully convolutional networks (ConvNets), the proposed method unifies hierarchically structured transformer encoder with Multi-Layer Perception (MLP) decoder in a Siamese network architecture to efficiently render multi-scale long-range details required for accurate CD. Experiments on two CD datasets show that the proposed end-to-end trainable ChangeFormer architecture achieves better CD performance than previous counterparts. Our code and pre-trained models are available at github.com/wgcban/ChangeFormer.
引用
收藏
页码：207 / 210
页数：4
相关论文
共 14 条
[1]  
Bandara W. G. C, 2021, arXiv
[2]   Remote Sensing Image Change Detection With Transformers [J].
Chen, Hao ;
Qi, Zipeng ;
Shi, Zhenwei .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2022, 60
[3]   A Spatial-Temporal Attention-Based Method and a New Dataset for Remote Sensing Image Change Detection [J].
Chen, Hao ;
Shi, Zhenwei .
REMOTE SENSING, 2020, 12 (10)
[4]  
Daudt RC, 2018, IEEE IMAGE PROC, P4063, DOI 10.1109/ICIP.2018.8451652
[5]  
Dosovitskiy A, 2021, Arxiv, DOI [arXiv:2010.11929, DOI 10.48550/ARXIV.2010.11929]
[6]   SNUNet-CD: A Densely Connected Siamese Network for Change Detection of VHR Images [J].
Fang, Sheng ;
Li, Kaiyu ;
Shao, Jinyuan ;
Li, Zhe .
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS, 2022, 19
[7]  
Bandara WGC, 2022, Arxiv, DOI arXiv:2204.08454
[8]  
Khan S., 2021, arXiv
[9]   Building Change Detection for Remote Sensing Images Using a Dual-Task Constrained Deep Siamese Convolutional Network Model [J].
Liu, Yi ;
Pang, Chao ;
Zhan, Zongqian ;
Zhang, Xiaomeng ;
Yang, Xue .
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS, 2021, 18 (05) :811-815
[10]   A Deeply Supervised Attention Metric-Based Network and an Open Aerial Image Dataset for Remote Sensing Change Detection [J].
Shi, Qian ;
Liu, Mengxi ;
Li, Shengchen ;
Liu, Xiaoping ;
Wang, Fei ;
Zhang, Liangpei .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2022, 60
12