Deep Feature Interactive Aggregation Network for Single Image Deraining
Single image deraining aims to remove rain streaks from a degraded input and reconstruct a high-quality image. In recent years, image processing tasks mostly applied a U-shaped architecture to capture rich contextual information. However, it is difficult to achieve long-range pixel dependencies beca...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2022-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9903649/ |
| _version_ | 1798028077928808448 |
|---|---|
| author | Shaoli Cao Liying Liu Li Zhao Yuewang Xu Jiawei Xu Xiaoqin Zhang |
| author_facet | Shaoli Cao Liying Liu Li Zhao Yuewang Xu Jiawei Xu Xiaoqin Zhang |
| author_sort | Shaoli Cao |
| collection | DOAJ |
| description | Single image deraining aims to remove rain streaks from a degraded input and reconstruct a high-quality image. In recent years, image processing tasks mostly applied a U-shaped architecture to capture rich contextual information. However, it is difficult to achieve long-range pixel dependencies because of the local receptive field of the convolution operation. In this paper, we propose a deep feature interactive aggregation network for single image deraining to enhance long-range dependencies among features and realize the interaction of information. To fully utilize high-level semantic features, we design a long-range dependency feature aggregation module to significantly improve the representational ability of the original U-shaped architecture. It aggregates multi-scale features and calculates the interactive attention of non-overlapping patches among feature maps. In addition, we adopt group normalization to retain the independence of each given image. It interacts with the information among features in an individual image and normalizes the channels of each group to weaken the correlation between batch data processing. Experimental results on widely acknowledged datasets also demonstrate the superiority of our proposed network over previous state-of-the-art methods. |
| first_indexed | 2024-04-11T19:02:13Z |
| format | Article |
| id | doaj.art-c19b4a031dff401db41aef19ca547401 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-04-11T19:02:13Z |
| publishDate | 2022-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-c19b4a031dff401db41aef19ca5474012022-12-22T04:07:58ZengIEEEIEEE Access2169-35362022-01-011010387210387910.1109/ACCESS.2022.32101909903649Deep Feature Interactive Aggregation Network for Single Image DerainingShaoli Cao0Liying Liu1Li Zhao2https://orcid.org/0000-0001-5787-2705Yuewang Xu3Jiawei Xu4https://orcid.org/0000-0002-1891-0542Xiaoqin Zhang5https://orcid.org/0000-0003-0958-7285College of Computer Science and Artificial Intelligence, Wenzhou University, Zhejiang, ChinaCollege of Computer Science and Artificial Intelligence, Wenzhou University, Zhejiang, ChinaCollege of Computer Science and Artificial Intelligence, Wenzhou University, Zhejiang, ChinaCollege of Computer Science and Artificial Intelligence, Wenzhou University, Zhejiang, ChinaCollege of Computer Science and Artificial Intelligence, Wenzhou University, Zhejiang, ChinaCollege of Computer Science and Artificial Intelligence, Wenzhou University, Zhejiang, ChinaSingle image deraining aims to remove rain streaks from a degraded input and reconstruct a high-quality image. In recent years, image processing tasks mostly applied a U-shaped architecture to capture rich contextual information. However, it is difficult to achieve long-range pixel dependencies because of the local receptive field of the convolution operation. In this paper, we propose a deep feature interactive aggregation network for single image deraining to enhance long-range dependencies among features and realize the interaction of information. To fully utilize high-level semantic features, we design a long-range dependency feature aggregation module to significantly improve the representational ability of the original U-shaped architecture. It aggregates multi-scale features and calculates the interactive attention of non-overlapping patches among feature maps. In addition, we adopt group normalization to retain the independence of each given image. It interacts with the information among features in an individual image and normalizes the channels of each group to weaken the correlation between batch data processing. Experimental results on widely acknowledged datasets also demonstrate the superiority of our proposed network over previous state-of-the-art methods.https://ieeexplore.ieee.org/document/9903649/Deep networkimage derainingtransformer |
| spellingShingle | Shaoli Cao Liying Liu Li Zhao Yuewang Xu Jiawei Xu Xiaoqin Zhang Deep Feature Interactive Aggregation Network for Single Image Deraining IEEE Access Deep network image deraining transformer |
| title | Deep Feature Interactive Aggregation Network for Single Image Deraining |
| title_full | Deep Feature Interactive Aggregation Network for Single Image Deraining |
| title_fullStr | Deep Feature Interactive Aggregation Network for Single Image Deraining |
| title_full_unstemmed | Deep Feature Interactive Aggregation Network for Single Image Deraining |
| title_short | Deep Feature Interactive Aggregation Network for Single Image Deraining |
| title_sort | deep feature interactive aggregation network for single image deraining |
| topic | Deep network image deraining transformer |
| url | https://ieeexplore.ieee.org/document/9903649/ |
| work_keys_str_mv | AT shaolicao deepfeatureinteractiveaggregationnetworkforsingleimagederaining AT liyingliu deepfeatureinteractiveaggregationnetworkforsingleimagederaining AT lizhao deepfeatureinteractiveaggregationnetworkforsingleimagederaining AT yuewangxu deepfeatureinteractiveaggregationnetworkforsingleimagederaining AT jiaweixu deepfeatureinteractiveaggregationnetworkforsingleimagederaining AT xiaoqinzhang deepfeatureinteractiveaggregationnetworkforsingleimagederaining |