Contrastive Learning for Compact Single Image Dehazing
arXiv:2104.09367
Abstract
Single image dehazing is a challenging ill-posed problem due to the severe information degeneration. However, existing deep learning based dehazing methods only adopt clear images as positive samples to guide the training of dehazing network while negative information is unexploited. Moreover, most of them focus on strengthening the dehazing network with an increase of depth and width, leading to a significant requirement of computation and memory. In this paper, we propose a novel contrastive regularization (CR) built upon contrastive learning to exploit both the information of hazy images and clear images as negative and positive samples, respectively. CR ensures that the restored image is pulled to closer to the clear image and pushed to far away from the hazy image in the representation space. Furthermore, considering trade-off between performance and memory storage, we develop a compact dehazing network based on autoencoder-like (AE) framework. It involves an adaptive mixup operation and a dynamic feature enhancement module, which can benefit from preserving information flow adaptively and expanding the receptive field to improve the network's transformation capability, respectively. We term our dehazing network with autoencoder and contrastive regularization as AECR-Net. The extensive experiments on synthetic and real-world datasets demonstrate that our AECR-Net surpass the state-of-the-art approaches. The code is released in https://github.com/GlassyWu/AECR-Net.
arXiv commentCVPR 2021 accepted paper. Code: https://github.com/GlassyWu/AECR-Net
References in corpus (15)
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- A Simple Framework for Contrastive Learning of Visual Representations
- mixup: Beyond Empirical Risk Minimization
- In Defense of the Triplet Loss for Person Re-Identification
- Data-Efficient Image Recognition with Contrastive Predictive Coding
- Contrastive Multiview Coding
- Deformable Convolutional Networks
- Deformable ConvNets v2: More Deformable, Better Results
- GridDehazeNet: Attention-Based Multi-Scale Network for Image Dehazing
- Contrastive Learning for Unpaired Image-to-Image Translation
- Densely Connected Pyramid Dehazing Network
- Multi-Scale Boosted Dehazing Network with Dense Feature Fusion
- Gated Fusion Network for Single Image Dehazing
- Learning Deformable Kernels for Image and Video Denoising
- Gated Context Aggregation Network for Image Dehazing and Deraining