Journal
INFORMATION FUSION
Volume 96, Issue -, Pages 297-311Publisher
ELSEVIER
DOI: 10.1016/j.inffus.2023.03.021
Keywords
Blind super-resolution; Self-supervised; Contrastive learning; Remote sensing image; Deep learning
Ask authors/readers for more resources
In recent years, single image super-resolution (SR) has attracted significant attention in the remote sensing area, with numerous methods making remarkable progress in this field. However, most of these methods assume a fixed known degradation process, limiting their performance when faced with real-world distribution deviations. To address this, blind image super-resolution for multiple and unknown degradations has been explored. This paper proposes a self-supervised degradation-guided adaptive network to bridge the domain gap between simulation and reality and achieves superior results compared to state-of-the-art methods.
Over the past few years, single image super-resolution (SR) has become a hotspot in the remote sensing area, and numerous methods have made remarkable progress in this fundamental task. However, they usually rely on the assumption that images suffer from a fixed known degradation process, e.g., bicubic downsampling. To save us from performance drop when real-world distribution deviates from the naive assumption, blind image super-resolution for multiple and unknown degradations has been explored. Nevertheless, the lack of a real-world dataset and the challenge of reasonable degradation estimation hinder us from moving forward. In this paper, a self-supervised degradation-guided adaptive network is proposed to mitigate the domain gap between simulation and reality. Firstly, the complicated degradations are characterized by robust representations in embedding space, which promote adaptability to the downstream SR network with degradation priors. Specifically, we incorporated contrastive learning to blind remote sensing image SR, which guides the reconstruction process by encouraging the positive representations (relevant information) while punishing the negatives. Besides, an effective dual-wise feature modulation network is proposed for feature adaptation. With the guide of degradation representations, we conduct modulation on feature and channel dimensions to transform the low-resolution features into the desired domain that is suitable for reconstructing high-resolution images. Extensive experiments on three mainstream datasets have demonstrated our superiority against state-of-the-art methods. Our source code can be found at https://github.com/XY-boy/DRSR
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