A<sup>2</sup>Net: Adjacent Aggregation Networks for Image Raindrop Removal
Existing methods for single images raindrop removal either have poor robustness or suffer from parameter burdens. In this paper, we propose a new Adjacent Aggregation Network (A<sup>2</sup>Net) with lightweight architectures to remove raindrops from single images. Instead of directly cas...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2020-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9045941/ |
| _version_ | 1818608433597775872 |
|---|---|
| author | Huangxing Lin Changxing Jing Yue Huang Xinghao Ding |
| author_facet | Huangxing Lin Changxing Jing Yue Huang Xinghao Ding |
| author_sort | Huangxing Lin |
| collection | DOAJ |
| description | Existing methods for single images raindrop removal either have poor robustness or suffer from parameter burdens. In this paper, we propose a new Adjacent Aggregation Network (A<sup>2</sup>Net) with lightweight architectures to remove raindrops from single images. Instead of directly cascading convolutional layers, we design an adjacent aggregation architecture to better fuse features for rich representations generation, which can lead to high quality images reconstruction. To further simplify the learning process, we utilize a problem-specific knowledge to force the network focus on the luminance channel in the YUV color space instead of all RGB channels. By combining adjacent aggregating operation with color space transformation, the proposed A<sup>2</sup>Net can achieve state-of-the-art performances on raindrop removal with significant parameters reduction. |
| first_indexed | 2024-12-16T14:42:34Z |
| format | Article |
| id | doaj.art-ad6cb2c7f6b6454dbcc7a654af9f2e32 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-16T14:42:34Z |
| publishDate | 2020-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-ad6cb2c7f6b6454dbcc7a654af9f2e322022-12-21T22:27:51ZengIEEEIEEE Access2169-35362020-01-018607696077910.1109/ACCESS.2020.29830879045941A<sup>2</sup>Net: Adjacent Aggregation Networks for Image Raindrop RemovalHuangxing Lin0https://orcid.org/0000-0001-6247-586XChangxing Jing1https://orcid.org/0000-0002-4721-4458Yue Huang2https://orcid.org/0000-0002-3913-9400Xinghao Ding3https://orcid.org/0000-0003-2288-5287School of Information, Xiamen University, Xiamen, ChinaSchool of Information, Xiamen University, Xiamen, ChinaSchool of Information, Xiamen University, Xiamen, ChinaSchool of Information, Xiamen University, Xiamen, ChinaExisting methods for single images raindrop removal either have poor robustness or suffer from parameter burdens. In this paper, we propose a new Adjacent Aggregation Network (A<sup>2</sup>Net) with lightweight architectures to remove raindrops from single images. Instead of directly cascading convolutional layers, we design an adjacent aggregation architecture to better fuse features for rich representations generation, which can lead to high quality images reconstruction. To further simplify the learning process, we utilize a problem-specific knowledge to force the network focus on the luminance channel in the YUV color space instead of all RGB channels. By combining adjacent aggregating operation with color space transformation, the proposed A<sup>2</sup>Net can achieve state-of-the-art performances on raindrop removal with significant parameters reduction.https://ieeexplore.ieee.org/document/9045941/Raindrop removalfeature aggregationYUV spacedeep learning |
| spellingShingle | Huangxing Lin Changxing Jing Yue Huang Xinghao Ding A<sup>2</sup>Net: Adjacent Aggregation Networks for Image Raindrop Removal IEEE Access Raindrop removal feature aggregation YUV space deep learning |
| title | A<sup>2</sup>Net: Adjacent Aggregation Networks for Image Raindrop Removal |
| title_full | A<sup>2</sup>Net: Adjacent Aggregation Networks for Image Raindrop Removal |
| title_fullStr | A<sup>2</sup>Net: Adjacent Aggregation Networks for Image Raindrop Removal |
| title_full_unstemmed | A<sup>2</sup>Net: Adjacent Aggregation Networks for Image Raindrop Removal |
| title_short | A<sup>2</sup>Net: Adjacent Aggregation Networks for Image Raindrop Removal |
| title_sort | a sup 2 sup net adjacent aggregation networks for image raindrop removal |
| topic | Raindrop removal feature aggregation YUV space deep learning |
| url | https://ieeexplore.ieee.org/document/9045941/ |
| work_keys_str_mv | AT huangxinglin asup2supnetadjacentaggregationnetworksforimageraindropremoval AT changxingjing asup2supnetadjacentaggregationnetworksforimageraindropremoval AT yuehuang asup2supnetadjacentaggregationnetworksforimageraindropremoval AT xinghaoding asup2supnetadjacentaggregationnetworksforimageraindropremoval |