EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

ATMformer: An Adaptive Token Merging Vision Transformer for Remote Sensing Image Scene Classification

Yi Niu, Zhuochen Song, Qingyu Luo, Guochao Chen, Mingming Ma, Fu Li

Remote Sensing · 15.02.2025 · 17 (4) · 660

DOI: 10.3390/rs17040660 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 15.02.2025. Verfasst von Yi Niu und 5 weiteren Autor:innen. Die Arbeit wurde bisher 15-mal zitiert (Crossref) und nennt 61 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, China Postdoctoral Science Foundation, Key Research and Development Program of Shaanxi.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In remote sensing image scene classification (RSISC) tasks, downsampling is crucial for reducing computational complexity and cache demands, enhancing the model’s generalization capability of deep neural networks. Traditional methods, such as regular fixed lattice approaches (pooling in CNN and token merging in transformers), often flatten distinguishing texture features, impacting classification performance. To address this, we propose an adaptive token merging transformer (ATMformer) that preserves essential local features by estimating the importance score of each token. This allows significant tokens to be isolated during merging, mitigating the risk of feature blurring. Our experiments on three widely used RSISC datasets (NWPU-RESISC45, Aerial Image Dataset, and EuroSAT) demonstrate that ATMformer achieves state-of-the-art performance across all datasets. These findings underscore the importance of effective downsampling techniques in maintaining classification accuracy while efficiently processing large-scale data.

Förderung

  • National Natural Science Foundation of China
  • China Postdoctoral Science Foundation
  • Key Research and Development Program of Shaanxi

Zitieren

APA

Niu, Y., Song, Z., Luo, Q., Chen, G., Ma, M., & Li, F. (2025). ATMformer: An Adaptive Token Merging Vision Transformer for Remote Sensing Image Scene Classification. Remote Sensing, 17(4), 660. https://doi.org/10.3390/rs17040660

Vancouver

Niu Y, Song Z, Luo Q, Chen G, Ma M, Li F. ATMformer: An Adaptive Token Merging Vision Transformer for Remote Sensing Image Scene Classification. Remote Sensing. 2025;17(4):660. doi:10.3390/rs17040660

BibTeX
@article{Niu2025,
  author = {Niu, Yi and Song, Zhuochen and Luo, Qingyu and Chen, Guochao and Ma, Mingming and Li, Fu},
  title = {ATMformer: An Adaptive Token Merging Vision Transformer for Remote Sensing Image Scene Classification},
  journal = {Remote Sensing},
  year = {2025},
  volume = {17},
  number = {4},
  pages = {660},
  doi = {10.3390/rs17040660}
}
Kostenlos starten →