URNet: A U-Shaped Residual Network for Lightweight Image Super-Resolution
It is extremely important and necessary for low computing power or portable devices to design more lightweight algorithms for image super-resolution (SR). Recently, most SR methods have achieved outstanding performance by sacrificing computational cost and memory storage, or vice versa. To address t...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-09-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/13/19/3848 |
| _version_ | 1827680638273060864 |
|---|---|
| author | Yuntao Wang Lin Zhao Liman Liu Huaifei Hu Wenbing Tao |
| author_facet | Yuntao Wang Lin Zhao Liman Liu Huaifei Hu Wenbing Tao |
| author_sort | Yuntao Wang |
| collection | DOAJ |
| description | It is extremely important and necessary for low computing power or portable devices to design more lightweight algorithms for image super-resolution (SR). Recently, most SR methods have achieved outstanding performance by sacrificing computational cost and memory storage, or vice versa. To address this problem, we introduce a lightweight U-shaped residual network (URNet) for fast and accurate image SR. Specifically, we propose a more effective feature distillation pyramid residual group (FDPRG) to extract features from low-resolution images. The FDPRG can effectively reuse the learned features with dense shortcuts and capture multi-scale information with a cascaded feature pyramid block. Based on the U-shaped structure, we utilize a step-by-step fusion strategy to improve the performance of feature fusion of different blocks. This strategy is different from the general SR methods which only use a single <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>C</mi><mi>o</mi><mi>n</mi><mi>c</mi><mi>a</mi><mi>t</mi></mrow></semantics></math></inline-formula> operation to fuse the features of all basic blocks. Moreover, a lightweight asymmetric residual non-local block is proposed to model the global context information and further improve the performance of SR. Finally, a high-frequency loss function is designed to alleviate smoothing image details caused by pixel-wise loss. Simultaneously, the proposed modules and high-frequency loss function can be easily plugged into multiple mature architectures to improve the performance of SR. Extensive experiments on multiple natural image datasets and remote sensing image datasets show the URNet achieves a better trade-off between image SR performance and model complexity against other state-of-the-art SR methods. |
| first_indexed | 2024-03-10T06:53:27Z |
| format | Article |
| id | doaj.art-cdb7650d9241443791296151b5cbec8e |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T06:53:27Z |
| publishDate | 2021-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-cdb7650d9241443791296151b5cbec8e2023-11-22T16:41:45ZengMDPI AGRemote Sensing2072-42922021-09-011319384810.3390/rs13193848URNet: A U-Shaped Residual Network for Lightweight Image Super-ResolutionYuntao Wang0Lin Zhao1Liman Liu2Huaifei Hu3Wenbing Tao4School of Biomedical Engineering, South-Central University for Nationalities, Wuhan 430074, ChinaNational Key Laboratory of Science and Technology on Multi-Spectral Information Processing, School of Artificial Intelligence and Automation, Huazhong University of Science and Technology, Wuhan 430074, ChinaSchool of Biomedical Engineering, South-Central University for Nationalities, Wuhan 430074, ChinaSchool of Biomedical Engineering, South-Central University for Nationalities, Wuhan 430074, ChinaNational Key Laboratory of Science and Technology on Multi-Spectral Information Processing, School of Artificial Intelligence and Automation, Huazhong University of Science and Technology, Wuhan 430074, ChinaIt is extremely important and necessary for low computing power or portable devices to design more lightweight algorithms for image super-resolution (SR). Recently, most SR methods have achieved outstanding performance by sacrificing computational cost and memory storage, or vice versa. To address this problem, we introduce a lightweight U-shaped residual network (URNet) for fast and accurate image SR. Specifically, we propose a more effective feature distillation pyramid residual group (FDPRG) to extract features from low-resolution images. The FDPRG can effectively reuse the learned features with dense shortcuts and capture multi-scale information with a cascaded feature pyramid block. Based on the U-shaped structure, we utilize a step-by-step fusion strategy to improve the performance of feature fusion of different blocks. This strategy is different from the general SR methods which only use a single <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>C</mi><mi>o</mi><mi>n</mi><mi>c</mi><mi>a</mi><mi>t</mi></mrow></semantics></math></inline-formula> operation to fuse the features of all basic blocks. Moreover, a lightweight asymmetric residual non-local block is proposed to model the global context information and further improve the performance of SR. Finally, a high-frequency loss function is designed to alleviate smoothing image details caused by pixel-wise loss. Simultaneously, the proposed modules and high-frequency loss function can be easily plugged into multiple mature architectures to improve the performance of SR. Extensive experiments on multiple natural image datasets and remote sensing image datasets show the URNet achieves a better trade-off between image SR performance and model complexity against other state-of-the-art SR methods.https://www.mdpi.com/2072-4292/13/19/3848single image super-resolutionlightweight image super-resolutionU-shaped residual networkdense shortcuteffective feature distillationhigh-frequency loss |
| spellingShingle | Yuntao Wang Lin Zhao Liman Liu Huaifei Hu Wenbing Tao URNet: A U-Shaped Residual Network for Lightweight Image Super-Resolution Remote Sensing single image super-resolution lightweight image super-resolution U-shaped residual network dense shortcut effective feature distillation high-frequency loss |
| title | URNet: A U-Shaped Residual Network for Lightweight Image Super-Resolution |
| title_full | URNet: A U-Shaped Residual Network for Lightweight Image Super-Resolution |
| title_fullStr | URNet: A U-Shaped Residual Network for Lightweight Image Super-Resolution |
| title_full_unstemmed | URNet: A U-Shaped Residual Network for Lightweight Image Super-Resolution |
| title_short | URNet: A U-Shaped Residual Network for Lightweight Image Super-Resolution |
| title_sort | urnet a u shaped residual network for lightweight image super resolution |
| topic | single image super-resolution lightweight image super-resolution U-shaped residual network dense shortcut effective feature distillation high-frequency loss |
| url | https://www.mdpi.com/2072-4292/13/19/3848 |
| work_keys_str_mv | AT yuntaowang urnetaushapedresidualnetworkforlightweightimagesuperresolution AT linzhao urnetaushapedresidualnetworkforlightweightimagesuperresolution AT limanliu urnetaushapedresidualnetworkforlightweightimagesuperresolution AT huaifeihu urnetaushapedresidualnetworkforlightweightimagesuperresolution AT wenbingtao urnetaushapedresidualnetworkforlightweightimagesuperresolution |