EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Enhanced Window-Based Self-Attention with Global and Multi-Scale Representations for Remote Sensing Image Super-Resolution

Yuting Lu, Shunzhou Wang, Binglu Wang, Xin Zhang, Xiaoxu Wang, Yongqiang Zhao

Remote Sensing · 02.08.2024 · 16 (15) · 2837

DOI: 10.3390/rs16152837 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 02.08.2024. Verfasst von Yuting Lu und 5 weiteren Autor:innen. Die Arbeit wurde bisher 22-mal zitiert (Crossref) und nennt 73 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Postdoctoral Science Foundation of China, Shaanxi Science Fund for Distinguished Young Scholars, Basic and Applied Basic Research Foundation of Guangdong Province.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Transformers have recently gained significant attention in low-level vision tasks, particularly for remote sensing image super-resolution (RSISR). The vanilla vision transformer aims to establish long-range dependencies between image patches. However, its global receptive field leads to a quadratic increase in computational complexity with respect to spatial size, rendering it inefficient for addressing RSISR tasks that involve processing large-sized images. In an effort to mitigate computational costs, recent studies have explored the utilization of local attention mechanisms, inspired by convolutional neural networks (CNNs), focusing on interactions between patches within small windows. Nevertheless, these approaches are naturally influenced by smaller participating receptive fields, and the utilization of fixed window sizes hinders their ability to perceive multi-scale information, consequently limiting model performance. To address these challenges, we propose a hierarchical transformer model named the Multi-Scale and Global Representation Enhancement-based Transformer (MSGFormer). We propose an efficient attention mechanism, Dual Window-based Self-Attention (DWSA), combining distributed and concentrated attention to balance computational complexity and the receptive field range. Additionally, we incorporated the Multi-scale Depth-wise Convolution Attention (MDCA) module, which is effective in capturing multi-scale features through multi-branch convolution. Furthermore, we developed a new Tracing-Back Structure (TBS), offering tracing-back mechanisms for both proposed attention modules to enhance their feature representation capability. Extensive experiments demonstrate that MSGFormer outperforms state-of-the-art methods on multiple public RSISR datasets by up to 0.11–0.55 dB.

Förderung

  • Postdoctoral Science Foundation of China
  • Shaanxi Science Fund for Distinguished Young Scholars
  • Basic and Applied Basic Research Foundation of Guangdong Province

Zitieren

APA

Lu, Y., Wang, S., Wang, B., Zhang, X., Wang, X., & Zhao, Y. (2024). Enhanced Window-Based Self-Attention with Global and Multi-Scale Representations for Remote Sensing Image Super-Resolution. Remote Sensing, 16(15), 2837. https://doi.org/10.3390/rs16152837

Vancouver

Lu Y, Wang S, Wang B, Zhang X, Wang X, Zhao Y. Enhanced Window-Based Self-Attention with Global and Multi-Scale Representations for Remote Sensing Image Super-Resolution. Remote Sensing. 2024;16(15):2837. doi:10.3390/rs16152837

BibTeX
@article{Lu2024,
  author = {Lu, Yuting and Wang, Shunzhou and Wang, Binglu and Zhang, Xin and Wang, Xiaoxu and Zhao, Yongqiang},
  title = {Enhanced Window-Based Self-Attention with Global and Multi-Scale Representations for Remote Sensing Image Super-Resolution},
  journal = {Remote Sensing},
  year = {2024},
  volume = {16},
  number = {15},
  pages = {2837},
  doi = {10.3390/rs16152837}
}
Kostenlos starten →