# LKD-Net: Large Kernel Convolution Network for Single Image Dehazing

arXiv: https://arxiv.org/abs/2209.01788 · HTML: https://mlsanity.com/paper/2209.01788/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- Builds on 24 of its 37 references that are in the map.
- First posted on arXiv 5 Sep 2022.

## Builds on (most cited first, 24 of 24)

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021)
- [Fully Convolutional Networks for Semantic Segmentation](https://mlsanity.com/paper/1411.4038/) (2014)
- [Squeeze-and-Excitation Networks](https://mlsanity.com/paper/1709.01507/) (2017)
- [A ConvNet for the 2020s](https://mlsanity.com/paper/2201.03545/) (2022)
- [Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network](https://mlsanity.com/paper/1609.04802/) (2016)
- [Restormer: Efficient Transformer for High-Resolution Image Restoration](https://mlsanity.com/paper/2111.09881/) (2021)
- [Understanding the Effective Receptive Field in Deep Convolutional Neural Networks](https://mlsanity.com/paper/1701.04128/) (2017)
- [Benchmarking Single Image Dehazing and Beyond](https://mlsanity.com/paper/1712.04143/) (2017)
- [Pay Attention to MLPs](https://mlsanity.com/paper/2105.08050/) (2021)
- [DehazeNet: An End-to-End System for Single Image Haze Removal](https://mlsanity.com/paper/1601.07661/) (2016)
- [FFA-Net: Feature Fusion Attention Network for Single Image Dehazing](https://mlsanity.com/paper/1911.07559/) (2019)
- [MAXIM: Multi-Axis MLP for Image Processing](https://mlsanity.com/paper/2201.02973/) (2022)
- [Dilated Residual Networks](https://mlsanity.com/paper/1705.09914/) (2017)
- [Vision Transformers for Single Image Dehazing](https://mlsanity.com/paper/2204.03883/) (2022)
- [Multi-Scale Boosted Dehazing Network with Dense Feature Fusion](https://mlsanity.com/paper/2004.13388/) (2020)
- [GridDehazeNet: Attention-Based Multi-Scale Network for Image Dehazing](https://mlsanity.com/paper/1908.03245/) (2019)
- [Contrastive Learning for Compact Single Image Dehazing](https://mlsanity.com/paper/2104.09367/) (2021)
- [Visual Attention Network](https://mlsanity.com/paper/2202.09741/) (2022)
- [Large Kernel Matters -- Improve Semantic Segmentation by Global Convolutional Network](https://mlsanity.com/paper/1703.02719/) (2017)
- [Scaling Up Your Kernels to 31x31: Revisiting Large Kernel Design in CNNs](https://mlsanity.com/paper/2203.06717/) (2022)
- [Gated Fusion Network for Single Image Dehazing](https://mlsanity.com/paper/1804.00213/) (2018)
- [You Only Look Yourself: Unsupervised and Untrained Single Image Dehazing Neural Network](https://mlsanity.com/paper/2006.16829/) (2020)
- [ConvMLP: Hierarchical Convolutional MLPs for Vision](https://mlsanity.com/paper/2109.04454/) (2021)

## Built on by (most cited first, 13 of 13)

- [SegMamba: Long-range Sequential Modeling Mamba For 3D Medical Image Segmentation](https://mlsanity.com/paper/2401.13560/) (2024)
- [UniRepLKNet: A Universal Perception Large-Kernel ConvNet for Audio, Video, Point Cloud, Time-Series and Image Recognition](https://mlsanity.com/paper/2311.15599/) (2023)
- [MixDehazeNet : Mix Structure Block For Image Dehazing Network](https://mlsanity.com/paper/2305.17654/) (2023)
- [Scaling Up Your Kernels: Large Kernel Design in ConvNets towards Universal Representations](https://mlsanity.com/paper/2410.08049/) (2024)
- [HDMba: Hyperspectral Remote Sensing Imagery Dehazing with State Space Model](https://mlsanity.com/paper/2406.05700/) (2024)
- [CasDyF-Net: Image Dehazing via Cascaded Dynamic Filters](https://mlsanity.com/paper/2409.08510/) (2024)
- [UDPNet: Unleashing Depth-based Priors for Robust Image Dehazing](https://mlsanity.com/paper/2601.06909/) (2026)
- [Towards Context-aware Convolutional Network for Image Restoration](https://mlsanity.com/paper/2412.11008/) (2024)
- [Cross Paradigm Representation and Alignment Transformer for Image Deraining](https://mlsanity.com/paper/2504.16455/) (2025)
- [Moir'e Zero: An Efficient and High-Performance Neural Architecture for Moir'e Removal](https://mlsanity.com/paper/2507.22407/) (2025)
- [VIViT: Variable-Input Vision Transformer Framework for 3D MR Image Segmentation](https://mlsanity.com/paper/2505.08693/) (2025)
- [DISK: Differentiable Sparse Kernel Complex for Efficient Spatially-Variant Convolution](https://mlsanity.com/paper/2512.04556/) (2025)
- [TTTIR: Unlocking Instance-Specific State Evolution via Test-Time Training for Image Restoration](https://mlsanity.com/paper/2609.26151/) (2026)

## Abstract (arXiv)

The deep convolutional neural networks (CNNs)-based single image dehazing methods have achieved significant success. The previous methods are devoted to improving the network's performance by increasing the network's depth and width. The current methods focus on increasing the convolutional kernel size to enhance its performance by benefiting from the larger receptive field. However, directly increasing the size of the convolutional kernel introduces a massive amount of computational overhead and parameters. Thus, a novel Large Kernel Convolution Dehaze Block (LKD Block) consisting of the Decomposition deep-wise Large Kernel Convolution Block (DLKCB) and the Channel Enhanced Feed-forward Network (CEFN) is devised in this paper. The designed DLKCB can split the deep-wise large kernel convolution into a smaller depth-wise convolution and a depth-wise dilated convolution without introducing massive parameters and computational overhead. Meanwhile, the designed CEFN incorporates a channel attention mechanism into Feed-forward Network to exploit significant channels and enhance robustness. By combining multiple LKD Blocks and Up-Down sampling modules, the Large Kernel Convolution Dehaze Network (LKD-Net) is conducted. The evaluation results demonstrate the effectiveness of the designed DLKCB and CEFN, and our LKD-Net outperforms the state-of-the-art. On the SOTS indoor dataset, our LKD-Net dramatically outperforms the Transformer-based method Dehamer with only 1.79% #Param and 48.9% FLOPs. The source code of our LKD-Net is available at https://github.com/SWU-CS-MediaLab/LKD-Net.