Implicit Euler ODE Networks for Single-Image Dehazing

Abstract

Deep convolutional neural networks (CNN) have been applied for image dehazing tasks, where the residual network (ResNet) is often adopted as the basic component to avoid the vanishing gradient problem. Recently, many works indicate that the ResNet can be considered as the explicit Euler forward approximation of an ordinary differential equation (ODE). In this paper, we extend the explicit forward approximation to the implicit backward counterpart, which can be realized via a recursive neural network, named IM-block. Given that, we propose an efficient end-to-end multi-level implicit network (MI-Net) for the single image dehazing problem. Moreover, multi-level fusing (MLF) mechanism and residual channel attention block (RCA-block) are adopted to boost performance of our network. Experiments on several dehazing benchmark datasets demonstrate that our method outperforms existing methods and achieves the state-of-the-art performance.

arXiv comment10pages, 10 figures, "for the associate project, see https://github.com/Jiawei-Shen?tab=repositories", submitted to CVPR workshop "vision for four seasons",

References in corpus (8)

• DehazeNet: An End-to-End System for Single Image Haze Removal
• Instance Normalization: The Missing Ingredient for Fast Stylization
• GridDehazeNet: Attention-Based Multi-Scale Network for Image Dehazing
• Densely Connected Pyramid Dehazing Network
• Reversible Architectures for Arbitrarily Deep Residual Neural Networks
• Gated Fusion Network for Single Image Dehazing
• Semantic Single-Image Dehazing
• Differential equations as models of deep neural networks