Linear Array Network for Low-light Image Enhancement

TL;DR

This paper introduces a Linear Array Self-attention mechanism and a corresponding network that effectively enhances low-light images by capturing global dependencies with reduced computational complexity, outperforming current state-of-the-art methods.

Contribution

The paper proposes LASA, a novel self-attention mechanism that efficiently models global relationships, and integrates it into LAN, a network that surpasses existing methods in low-light image enhancement.

Findings

LAN outperforms SOTA methods in RGB and RAW low-light enhancement

LAN achieves higher quality with fewer parameters

LASA reduces computational complexity of global feature modeling

Abstract

Convolution neural networks (CNNs) based methods have dominated the low-light image enhancement tasks due to their outstanding performance. However, the convolution operation is based on a local sliding window mechanism, which is difficult to construct the long-range dependencies of the feature maps. Meanwhile, the self-attention based global relationship aggregation methods have been widely used in computer vision, but these methods are difficult to handle high-resolution images because of the high computational complexity. To solve this problem, this paper proposes a Linear Array Self-attention (LASA) mechanism, which uses only two 2-D feature encodings to construct 3-D global weights and then refines feature maps generated by convolution layers. Based on LASA, Linear Array Network (LAN) is proposed, which is superior to the existing state-of-the-art (SOTA) methods in both RGB and RAW based low-light enhancement tasks with a smaller amount of parameters. The code is released in https://github.com/cuiziteng/LASA_enhancement.