English

Efficient Non-Local Contrastive Attention for Image Super-Resolution

Computer Vision and Pattern Recognition 2022-03-11 v2 Artificial Intelligence

Abstract

Non-Local Attention (NLA) brings significant improvement for Single Image Super-Resolution (SISR) by leveraging intrinsic feature correlation in natural images. However, NLA gives noisy information large weights and consumes quadratic computation resources with respect to the input size, limiting its performance and application. In this paper, we propose a novel Efficient Non-Local Contrastive Attention (ENLCA) to perform long-range visual modeling and leverage more relevant non-local features. Specifically, ENLCA consists of two parts, Efficient Non-Local Attention (ENLA) and Sparse Aggregation. ENLA adopts the kernel method to approximate exponential function and obtains linear computation complexity. For Sparse Aggregation, we multiply inputs by an amplification factor to focus on informative features, yet the variance of approximation increases exponentially. Therefore, contrastive learning is applied to further separate relevant and irrelevant features. To demonstrate the effectiveness of ENLCA, we build an architecture called Efficient Non-Local Contrastive Network (ENLCN) by adding a few of our modules in a simple backbone. Extensive experimental results show that ENLCN reaches superior performance over state-of-the-art approaches on both quantitative and qualitative evaluations.

Keywords

Cite

@article{arxiv.2201.03794,
  title  = {Efficient Non-Local Contrastive Attention for Image Super-Resolution},
  author = {Bin Xia and Yucheng Hang and Yapeng Tian and Wenming Yang and Qingmin Liao and Jie Zhou},
  journal= {arXiv preprint arXiv:2201.03794},
  year   = {2022}
}

Comments

Code is available at https://github.com/Zj-BinXia/ENLCA

R2 v1 2026-06-24T08:46:00.955Z