ChangeViT: Unleashing plain vision transformers for change detection in remote sensing images
-
-
DOI number:
10.1016/j.patcog.2025.112539
-
Affiliation of Author(s):
Wuhan Univ, State Key Lab Informat Engn Surveying Mapping & Re, Wuhan, Peoples R China
-
-
Journal:
PATTERN RECOGNITION
-
-
Funded by:
This work was supported by the Fundamental Research Funds for the Central Universities (2042024kf003
-
Key Words:
Change detection,Vision transformer,Remote sensing images
-
Abstract:
Change detection in remote sensing images is essential for monitoring environmental changes on Earth's surface. Although vision transformers (ViTs), particularly plain ViTs pre-trained on large-scale datasets and endowed with extensive knowledge and strong comprehension capabilities, have achieved remarkable success as backbones in numerous computer vision applications, they remain underutilized in change detection tasks. In this domain, convolutional neural networks (CNNs) continue to dominate due to their powerful feature extraction capabilities. In this paper, our study uncovers ViTs' uniqu
-
-
Co-author:
Haiyan, Zheng,Huang, Qimin,Cheng, Xiaohu
-
-
Indexed by:
Journal paper
-
Correspondence Author:
Shao, Zhenfeng
-
-
-
-
Document Type:
Article
-
Volume:
172
-
-
-
-
ISSN No.:
0031-3203
-
Translation or Not:
no
-
-
Date of Publication:
2026-04-01
-
-