Frequency-Spatial Interaction Driven Network for Low-Light Image Enhancement

2025-10-25

AI Summary

To address insufficient exploitation of frequency-domain and spatial-domain information, as well as limited cross-scale feature interaction in low-light image enhancement (LLIE), this paper proposes a two-stage frequency-spatial collaborative network: the first stage restores global brightness, while the second reconstructs local details. We introduce a novel frequency-spatial interaction module—the first to enable bidirectional feature mapping between the Fourier and spatial domains—and incorporate cross-stage, cross-scale information exchange to jointly optimize amplitude and phase recovery. Our method achieves significant PSNR/SSIM improvements on benchmarks including LOL-Real and LSRW-Huawei, while maintaining efficient inference. The core contribution lies in establishing a learnable, bidirectional frequency-spatial interaction paradigm that simultaneously balances enhancement quality, structural fidelity, and computational efficiency.

Technology Category

• Computer Vision: Multi-modal Vision
• Search and Optimization: Learning to Search
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Large language models for search
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Low-light image enhancement (LLIE) aims at improving the perception or interpretability of an image captured in an environment with poor illumination. With the advent of deep learning, the LLIE technique has achieved significant breakthroughs. However, existing LLIE methods either ignore the important role of frequency domain information or fail to effectively promote the propagation and flow of information, limiting the LLIE performance. In this paper, we develop a novel frequency-spatial interaction-driven network (FSIDNet) for LLIE based on two-stage architecture. To be specific, the first stage is designed to restore the amplitude of low-light images to improve the lightness, and the second stage devotes to restore phase information to refine fine-grained structures. Considering that Frequency domain and spatial domain information are complementary and both favorable for LLIE, we further develop two frequency-spatial interaction blocks which mutually amalgamate the complementary spatial and frequency information to enhance the capability of the model. In addition, we construct the Information Exchange Module (IEM) to associate two stages by adequately incorporating cross-stage and cross-scale features to effectively promote the propagation and flow of information in the two-stage network structure. Finally, we conduct experiments on several widely used benchmark datasets (i.e., LOL-Real, LSRW-Huawei, etc.), which demonstrate that our method achieves the excellent performance in terms of visual results and quantitative metrics while preserving good model efficiency.

Problem

• Enhancing low-light images using frequency-spatial interaction
• Restoring amplitude and phase information in two stages
• Improving information flow across stages and scales

Innovation

• Two-stage architecture restores amplitude and phase
• Frequency-spatial interaction blocks amalgamate complementary information
• Information Exchange Module promotes cross-stage feature propagation