Degradation-Aware Transformer for Single Image Deraining
The crux of image deraining originates from recognizing various rain patterns. Most existing methods employ low-level spatial or sequential information to reconstruct the rain-free background image. However, due to the lack of sufficiently capturing long-term contextual relation between pixels, thes...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2023-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10237180/ |
| _version_ | 1797685380356505600 |
|---|---|
| author | Peijun Zhao Tongjun Wang |
| author_facet | Peijun Zhao Tongjun Wang |
| author_sort | Peijun Zhao |
| collection | DOAJ |
| description | The crux of image deraining originates from recognizing various rain patterns. Most existing methods employ low-level spatial or sequential information to reconstruct the rain-free background image. However, due to the lack of sufficiently capturing long-term contextual relation between pixels, these methods often lead to incompletely modeling rain patterns and visible rain residues remained. In this paper, we propose a novel Degradation-Aware Transformer (DAT), which leverages a multi-level contrastive learning to obtain discriminative degradation representations by a degradation-aware model. Based on this, we also design a degradation-aware self-attention mechanism to improve the restoration performance on diverse rain patterns. Benefiting from the developed self-attention mechanism, DAT is able to capture long-term relations between pixels as well as completely modeling the rain degradation patterns. Extensive experiments demonstrate that our DAT is able to achieve the state-of-the-art performance of single image deraining in terms of both qualitatively visual comparison and quantitative comparison with other baseline methods on benchmark datasets. |
| first_indexed | 2024-03-12T00:44:21Z |
| format | Article |
| id | doaj.art-9a20f74eafd84154b093bc6f9942db95 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-03-12T00:44:21Z |
| publishDate | 2023-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-9a20f74eafd84154b093bc6f9942db952023-09-14T23:00:37ZengIEEEIEEE Access2169-35362023-01-0111972749728310.1109/ACCESS.2023.331113810237180Degradation-Aware Transformer for Single Image DerainingPeijun Zhao0https://orcid.org/0009-0009-1045-0236Tongjun Wang1School of Information Engineering, Xinyang Agriculture and Forestry University, Xinyang, ChinaSchool of Information Engineering, Xinyang Agriculture and Forestry University, Xinyang, ChinaThe crux of image deraining originates from recognizing various rain patterns. Most existing methods employ low-level spatial or sequential information to reconstruct the rain-free background image. However, due to the lack of sufficiently capturing long-term contextual relation between pixels, these methods often lead to incompletely modeling rain patterns and visible rain residues remained. In this paper, we propose a novel Degradation-Aware Transformer (DAT), which leverages a multi-level contrastive learning to obtain discriminative degradation representations by a degradation-aware model. Based on this, we also design a degradation-aware self-attention mechanism to improve the restoration performance on diverse rain patterns. Benefiting from the developed self-attention mechanism, DAT is able to capture long-term relations between pixels as well as completely modeling the rain degradation patterns. Extensive experiments demonstrate that our DAT is able to achieve the state-of-the-art performance of single image deraining in terms of both qualitatively visual comparison and quantitative comparison with other baseline methods on benchmark datasets.https://ieeexplore.ieee.org/document/10237180/Image derainingdegradation representationcontrastive learninggenerative model |
| spellingShingle | Peijun Zhao Tongjun Wang Degradation-Aware Transformer for Single Image Deraining IEEE Access Image deraining degradation representation contrastive learning generative model |
| title | Degradation-Aware Transformer for Single Image Deraining |
| title_full | Degradation-Aware Transformer for Single Image Deraining |
| title_fullStr | Degradation-Aware Transformer for Single Image Deraining |
| title_full_unstemmed | Degradation-Aware Transformer for Single Image Deraining |
| title_short | Degradation-Aware Transformer for Single Image Deraining |
| title_sort | degradation aware transformer for single image deraining |
| topic | Image deraining degradation representation contrastive learning generative model |
| url | https://ieeexplore.ieee.org/document/10237180/ |
| work_keys_str_mv | AT peijunzhao degradationawaretransformerforsingleimagederaining AT tongjunwang degradationawaretransformerforsingleimagederaining |