Linear Array Network for Low-light Image Enhancement
This work improves low-light image enhancement for computer vision applications by introducing a more efficient attention mechanism, though it is incremental as it builds on existing CNN and self-attention approaches.
The paper tackles the problem of low-light image enhancement by proposing a Linear Array Self-attention (LASA) mechanism to address the limitations of CNNs in capturing long-range dependencies and the high computational complexity of self-attention methods, resulting in a Linear Array Network (LAN) that outperforms state-of-the-art methods in RGB and RAW tasks with fewer parameters.
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.