Journal
2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021
Volume -, Issue -, Pages 10546-10555Publisher
IEEE COMPUTER SOC
DOI: 10.1109/CVPR46437.2021.01041
Keywords
-
Funding
- National Natural Science Foundation of China [61772524, 61876161, 61972157]
- National Key Research and Development Program of China [2020AAA0108301]
- Natural Science Foundation of Shanghai [20ZR1417700]
- CAAI-Huawei MindSpore Open Fund
- Research Program of Zhejiang Lab [2019KD0AC02]
Ask authors/readers for more resources
This study proposes a novel contrastive regularization method for single image dehazing. By utilizing contrastive learning, clear images are used as positive samples and hazy images as negative samples, ensuring that the restored image is closer to the clear image and away from the hazy image in the representation space. Additionally, a compact dehazing network based on an autoencoder framework is developed to balance performance and memory storage, incorporating adaptive mixup operation and dynamic feature enhancement module.
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.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available