Decoupling Induction and Multi-Order Attention Drop-Out Gating Based Joint Motion Deblurring and Image Super-Resolution
Resolution decrease and motion blur are two typical image degradation processes that are usually addressed by deep networks, specifically convolutional neural networks (CNNs). However, since real images are usually obtained through multiple degradations, the vast majority of current CNN methods that...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-05-01
|
| Series: | Mathematics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7390/10/11/1837 |
| _version_ | 1797492601178292224 |
|---|---|
| author | Yuezhong Chu Xuefeng Zhang Heng Liu |
| author_facet | Yuezhong Chu Xuefeng Zhang Heng Liu |
| author_sort | Yuezhong Chu |
| collection | DOAJ |
| description | Resolution decrease and motion blur are two typical image degradation processes that are usually addressed by deep networks, specifically convolutional neural networks (CNNs). However, since real images are usually obtained through multiple degradations, the vast majority of current CNN methods that employ a single degradation process inevitably need to be improved to account for multiple degradation effects. In this work, motivated by degradation decoupling and multiple-order attention drop-out gating, we propose a joint deep recovery model to efficiently address motion blur and resolution reduction simultaneously. Our degradation decoupling style improves the continence and the efficiency of model construction and training. Moreover, the proposed multi-order attention mechanism comprehensively and hierarchically extracts multiple attention features and fuses them properly by drop-out gating. The proposed approach is evaluated using diverse benchmark datasets including natural and synthetic images. The experimental results show that our proposed method can efficiently complete joint motion blur and image super-resolution (SR). |
| first_indexed | 2024-03-10T01:06:02Z |
| format | Article |
| id | doaj.art-045f72545b0c4a8f9ec79059fa099d10 |
| institution | Directory Open Access Journal |
| issn | 2227-7390 |
| language | English |
| last_indexed | 2024-03-10T01:06:02Z |
| publishDate | 2022-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Mathematics |
| spelling | doaj.art-045f72545b0c4a8f9ec79059fa099d102023-11-23T14:25:23ZengMDPI AGMathematics2227-73902022-05-011011183710.3390/math10111837Decoupling Induction and Multi-Order Attention Drop-Out Gating Based Joint Motion Deblurring and Image Super-ResolutionYuezhong Chu0Xuefeng Zhang1Heng Liu2School of Computer Science and Technology, Anhui University of Technology, Ma’anshan 243002, ChinaSchool of Computer Science and Technology, Anhui University of Technology, Ma’anshan 243002, ChinaSchool of Computer Science and Technology, Anhui University of Technology, Ma’anshan 243002, ChinaResolution decrease and motion blur are two typical image degradation processes that are usually addressed by deep networks, specifically convolutional neural networks (CNNs). However, since real images are usually obtained through multiple degradations, the vast majority of current CNN methods that employ a single degradation process inevitably need to be improved to account for multiple degradation effects. In this work, motivated by degradation decoupling and multiple-order attention drop-out gating, we propose a joint deep recovery model to efficiently address motion blur and resolution reduction simultaneously. Our degradation decoupling style improves the continence and the efficiency of model construction and training. Moreover, the proposed multi-order attention mechanism comprehensively and hierarchically extracts multiple attention features and fuses them properly by drop-out gating. The proposed approach is evaluated using diverse benchmark datasets including natural and synthetic images. The experimental results show that our proposed method can efficiently complete joint motion blur and image super-resolution (SR).https://www.mdpi.com/2227-7390/10/11/1837motion deblurringimage super-resolutionmulti-order attentiongated learningdecoupling |
| spellingShingle | Yuezhong Chu Xuefeng Zhang Heng Liu Decoupling Induction and Multi-Order Attention Drop-Out Gating Based Joint Motion Deblurring and Image Super-Resolution Mathematics motion deblurring image super-resolution multi-order attention gated learning decoupling |
| title | Decoupling Induction and Multi-Order Attention Drop-Out Gating Based Joint Motion Deblurring and Image Super-Resolution |
| title_full | Decoupling Induction and Multi-Order Attention Drop-Out Gating Based Joint Motion Deblurring and Image Super-Resolution |
| title_fullStr | Decoupling Induction and Multi-Order Attention Drop-Out Gating Based Joint Motion Deblurring and Image Super-Resolution |
| title_full_unstemmed | Decoupling Induction and Multi-Order Attention Drop-Out Gating Based Joint Motion Deblurring and Image Super-Resolution |
| title_short | Decoupling Induction and Multi-Order Attention Drop-Out Gating Based Joint Motion Deblurring and Image Super-Resolution |
| title_sort | decoupling induction and multi order attention drop out gating based joint motion deblurring and image super resolution |
| topic | motion deblurring image super-resolution multi-order attention gated learning decoupling |
| url | https://www.mdpi.com/2227-7390/10/11/1837 |
| work_keys_str_mv | AT yuezhongchu decouplinginductionandmultiorderattentiondropoutgatingbasedjointmotiondeblurringandimagesuperresolution AT xuefengzhang decouplinginductionandmultiorderattentiondropoutgatingbasedjointmotiondeblurringandimagesuperresolution AT hengliu decouplinginductionandmultiorderattentiondropoutgatingbasedjointmotiondeblurringandimagesuperresolution |