EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Geometry-Aware Attention with a Shared Second-Order Feature Matrix for Lightweight Image Super-Resolution

Xudan Zhou, Hanjiang Lin, Huiying Xu, Xinpeng Zhuang, Hongdan Gu, Wenhao Li, Zhengdong Chen, Xinzhong Zhu

Electronics · 29.10.2026 · 15 (21) · 5020

DOI: 10.3390/electronics15215020 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Electronics, online erschienen am 29.10.2026. Verfasst von Xudan Zhou und 7 weiteren Autor:innen. Die Arbeit nennt 47 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Zhejiang Province Leading Geese Plan, Zhejiang Province Province-Land Synergy Program.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Existing lightweight single-image super-resolution (SISR) methods often rely on standard dot-product attention when explicit token-to-token similarity is modeled. Although the query and key projections are learned, the similarity in the projected coordinates is still evaluated by the Euclidean inner product. In this work, we propose GAASR, a lightweight SISR network that augments window-based attention with a shared, data-conditioned second-order feature matrix. For each attention layer, an uncentered second-order descriptor is pooled across the windows and heads of the current forward pass and combined with a learnable base matrix. The resulting symmetric matrix is shared within that layer and is used in two complementary ways: to linearly transform query features and to precondition the token-aligned query–key residual through a linear solve. A trace-derived scalar then controls the contribution of the residual correction before the standard dot-product attention is evaluated. This design introduces only small feature-dimension-dependent overhead and preserves the dominant quadratic dependence of window attention on the number of tokens. The proposed mechanism is formulated using symmetric matrix transformations and linear-system solving while retaining the standard dot-product attention structure. Experiments on multiple standard SISR benchmarks show consistent reconstruction gains under comparable parameter counts and reported computational budgets, with noticeable improvements on texture-rich and high-frequency regions.

Förderung

  • National Natural Science Foundation of China
  • Zhejiang Province Leading Geese Plan
  • Zhejiang Province Province-Land Synergy Program

Zitieren

APA

Zhou, X., Lin, H., Xu, H., Zhuang, X., Gu, H., Li, W., Chen, Z., & Zhu, X. (2026). Geometry-Aware Attention with a Shared Second-Order Feature Matrix for Lightweight Image Super-Resolution. Electronics, 15(21), 5020. https://doi.org/10.3390/electronics15215020

Vancouver

Zhou X, Lin H, Xu H, Zhuang X, Gu H, Li W, et al. Geometry-Aware Attention with a Shared Second-Order Feature Matrix for Lightweight Image Super-Resolution. Electronics. 2026;15(21):5020. doi:10.3390/electronics15215020

BibTeX
@article{Zhou2026,
  author = {Zhou, Xudan and Lin, Hanjiang and Xu, Huiying and Zhuang, Xinpeng and Gu, Hongdan and Li, Wenhao and Chen, Zhengdong and Zhu, Xinzhong},
  title = {Geometry-Aware Attention with a Shared Second-Order Feature Matrix for Lightweight Image Super-Resolution},
  journal = {Electronics},
  year = {2026},
  volume = {15},
  number = {21},
  pages = {5020},
  doi = {10.3390/electronics15215020}
}
Kostenlos starten →