EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Efficient Hierarchical Non-Local Retrieval for Arbitrary-Scale Image Super-Resolution

You Mou, Zhiyong Qi, Xin Su, Zongyi Peng, Luguang Ran, Li Li

Applied Sciences · 16.11.2026 · 16 (22) · 11359

DOI: 10.3390/app162211359 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Applied Sciences, online erschienen am 16.11.2026. Verfasst von You Mou und 5 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: China Yangtze Power Co.,Ltd..

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Implicit image functions enable a single model to perform arbitrary-scale image super-resolution. However, most existing methods rely mainly on local features near each query coordinate, limiting their ability to exploit repeated textures and long-range structural information. Dense non-local matching can complement local features but requires expensive exhaustive feature comparisons on high-resolution feature maps. To address this problem, we propose Sparse Attention with Coarse-to-Fine Aggregation (SACA), an efficient hierarchical non-local retrieval method for arbitrary-scale image super-resolution. SACA performs global similarity search only at the coarsest feature level to identify potentially relevant regions, and progressively propagates and re-evaluates the selected candidates at finer levels. A local refinement module further adjusts the sampling locations around the retrieved candidates before feature aggregation. Experiments on DIV2K, Urban100, and B100 demonstrate competitive reconstruction performance at both in-scale and out-of-scale factors. On Urban100, SACA achieves consistent improvements over LTE, with a peak signal-to-noise ratio (PSNR) gain of 0.1693 dB at ×8. Meanwhile, the proposed coarse-to-fine retrieval reduces the computational cost by up to 82.4% compared with exhaustive fine-level matching over the evaluated input sizes. These results demonstrate that SACA provides an effective balance between reconstruction quality and computational efficiency.

Förderung

  • China Yangtze Power Co.,Ltd.

Zitieren

APA

Mou, Y., Qi, Z., Su, X., Peng, Z., Ran, L., & Li, L. (2026). Efficient Hierarchical Non-Local Retrieval for Arbitrary-Scale Image Super-Resolution. Applied Sciences, 16(22), 11359. https://doi.org/10.3390/app162211359

Vancouver

Mou Y, Qi Z, Su X, Peng Z, Ran L, Li L. Efficient Hierarchical Non-Local Retrieval for Arbitrary-Scale Image Super-Resolution. Applied Sciences. 2026;16(22):11359. doi:10.3390/app162211359

BibTeX
@article{Mou2026,
  author = {Mou, You and Qi, Zhiyong and Su, Xin and Peng, Zongyi and Ran, Luguang and Li, Li},
  title = {Efficient Hierarchical Non-Local Retrieval for Arbitrary-Scale Image Super-Resolution},
  journal = {Applied Sciences},
  year = {2026},
  volume = {16},
  number = {22},
  pages = {11359},
  doi = {10.3390/app162211359}
}
Kostenlos starten →