Rethinking Performance Gains in Image Dehazing Networks
This work addresses the need for efficient and effective image dehazing in low-level vision, though it is incremental as it builds on existing U-Net architecture.
The authors tackled the problem of unclear mechanisms for performance gains in image dehazing networks by making minimal modifications to U-Net, resulting in gUNet, which achieved superior performance to state-of-the-art methods on multiple datasets with reduced overhead.
Image dehazing is an active topic in low-level vision, and many image dehazing networks have been proposed with the rapid development of deep learning. Although these networks' pipelines work fine, the key mechanism to improving image dehazing performance remains unclear. For this reason, we do not target to propose a dehazing network with fancy modules; rather, we make minimal modifications to popular U-Net to obtain a compact dehazing network. Specifically, we swap out the convolutional blocks in U-Net for residual blocks with the gating mechanism, fuse the feature maps of main paths and skip connections using the selective kernel, and call the resulting U-Net variant gUNet. As a result, with a significantly reduced overhead, gUNet is superior to state-of-the-art methods on multiple image dehazing datasets. Finally, we verify these key designs to the performance gain of image dehazing networks through extensive ablation studies.