An Omni-Dimensional Dynamic Convolutional Network for Single-Image Super-Resolution Tasks
The goal of single-image super-resolution (SISR) tasks is to generate high-definition images from low-quality inputs, with practical uses spanning healthcare diagnostics, aerial imaging, and surveillance systems. Although cnns have considerably improved image reconstruction quality, existing methods...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-07-01
|
| Series: | Mathematics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7390/13/15/2388 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849239762986074112 |
|---|---|
| author | Xi Chen Ziang Wu Weiping Zhang Tingting Bi Chunwei Tian |
| author_facet | Xi Chen Ziang Wu Weiping Zhang Tingting Bi Chunwei Tian |
| author_sort | Xi Chen |
| collection | DOAJ |
| description | The goal of single-image super-resolution (SISR) tasks is to generate high-definition images from low-quality inputs, with practical uses spanning healthcare diagnostics, aerial imaging, and surveillance systems. Although cnns have considerably improved image reconstruction quality, existing methods still face limitations, including inadequate restoration of high-frequency details, high computational complexity, and insufficient adaptability to complex scenes. To address these challenges, we propose an Omni-dimensional Dynamic Convolutional Network (ODConvNet) tailored for SISR tasks. Specifically, ODConvNet comprises four key components: a Feature Extraction Block (FEB) that captures low-level spatial features; an Omni-dimensional Dynamic Convolution Block (DCB), which utilizes a multidimensional attention mechanism to dynamically reweight convolution kernels across spatial, channel, and kernel dimensions, thereby enhancing feature expressiveness and context modeling; a Deep Feature Extraction Block (DFEB) that stacks multiple convolutional layers with residual connections to progressively extract and fuse high-level features; and a Reconstruction Block (RB) that employs subpixel convolution to upscale features and refine the final HR output. This mechanism significantly enhances feature extraction and effectively captures rich contextual information. Additionally, we employ an improved residual network structure combined with a refined Charbonnier loss function to alleviate gradient vanishing and exploding to enhance the robustness of model training. Extensive experiments conducted on widely used benchmark datasets, including DIV2K, Set5, Set14, B100, and Urban100, demonstrate that, compared with existing deep learning-based SR methods, our ODConvNet method improves Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM), and the visual quality of SR images is also improved. Ablation studies further validate the effectiveness and contribution of each component in our network. The proposed ODConvNet offers an effective, flexible, and efficient solution for the SISR task and provides promising directions for future research. |
| format | Article |
| id | doaj-art-d229b13f7a0143f1bbd5229ecaa68673 |
| institution | Kabale University |
| issn | 2227-7390 |
| language | English |
| publishDate | 2025-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Mathematics |
| spelling | doaj-art-d229b13f7a0143f1bbd5229ecaa686732025-08-20T04:00:50ZengMDPI AGMathematics2227-73902025-07-011315238810.3390/math13152388An Omni-Dimensional Dynamic Convolutional Network for Single-Image Super-Resolution TasksXi Chen0Ziang Wu1Weiping Zhang2Tingting Bi3Chunwei Tian4School of Software, Northwestern Polytechnical University, Xi’an 710129, ChinaSchool of Software, Northwestern Polytechnical University, Xi’an 710129, ChinaSchool of Software, Northwestern Polytechnical University, Xi’an 710129, ChinaSchool of Computing and Information Systems, University of Melbourne, Parkville 3010, AustraliaShenzhen Research Institute of Northwestern Polytechnical University, Shenzhen 518057, ChinaThe goal of single-image super-resolution (SISR) tasks is to generate high-definition images from low-quality inputs, with practical uses spanning healthcare diagnostics, aerial imaging, and surveillance systems. Although cnns have considerably improved image reconstruction quality, existing methods still face limitations, including inadequate restoration of high-frequency details, high computational complexity, and insufficient adaptability to complex scenes. To address these challenges, we propose an Omni-dimensional Dynamic Convolutional Network (ODConvNet) tailored for SISR tasks. Specifically, ODConvNet comprises four key components: a Feature Extraction Block (FEB) that captures low-level spatial features; an Omni-dimensional Dynamic Convolution Block (DCB), which utilizes a multidimensional attention mechanism to dynamically reweight convolution kernels across spatial, channel, and kernel dimensions, thereby enhancing feature expressiveness and context modeling; a Deep Feature Extraction Block (DFEB) that stacks multiple convolutional layers with residual connections to progressively extract and fuse high-level features; and a Reconstruction Block (RB) that employs subpixel convolution to upscale features and refine the final HR output. This mechanism significantly enhances feature extraction and effectively captures rich contextual information. Additionally, we employ an improved residual network structure combined with a refined Charbonnier loss function to alleviate gradient vanishing and exploding to enhance the robustness of model training. Extensive experiments conducted on widely used benchmark datasets, including DIV2K, Set5, Set14, B100, and Urban100, demonstrate that, compared with existing deep learning-based SR methods, our ODConvNet method improves Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM), and the visual quality of SR images is also improved. Ablation studies further validate the effectiveness and contribution of each component in our network. The proposed ODConvNet offers an effective, flexible, and efficient solution for the SISR task and provides promising directions for future research.https://www.mdpi.com/2227-7390/13/15/2388single image super-resolutionomni-dimensional dynamic convolutionresidual networksCharbonnier loss |
| spellingShingle | Xi Chen Ziang Wu Weiping Zhang Tingting Bi Chunwei Tian An Omni-Dimensional Dynamic Convolutional Network for Single-Image Super-Resolution Tasks Mathematics single image super-resolution omni-dimensional dynamic convolution residual networks Charbonnier loss |
| title | An Omni-Dimensional Dynamic Convolutional Network for Single-Image Super-Resolution Tasks |
| title_full | An Omni-Dimensional Dynamic Convolutional Network for Single-Image Super-Resolution Tasks |
| title_fullStr | An Omni-Dimensional Dynamic Convolutional Network for Single-Image Super-Resolution Tasks |
| title_full_unstemmed | An Omni-Dimensional Dynamic Convolutional Network for Single-Image Super-Resolution Tasks |
| title_short | An Omni-Dimensional Dynamic Convolutional Network for Single-Image Super-Resolution Tasks |
| title_sort | omni dimensional dynamic convolutional network for single image super resolution tasks |
| topic | single image super-resolution omni-dimensional dynamic convolution residual networks Charbonnier loss |
| url | https://www.mdpi.com/2227-7390/13/15/2388 |
| work_keys_str_mv | AT xichen anomnidimensionaldynamicconvolutionalnetworkforsingleimagesuperresolutiontasks AT ziangwu anomnidimensionaldynamicconvolutionalnetworkforsingleimagesuperresolutiontasks AT weipingzhang anomnidimensionaldynamicconvolutionalnetworkforsingleimagesuperresolutiontasks AT tingtingbi anomnidimensionaldynamicconvolutionalnetworkforsingleimagesuperresolutiontasks AT chunweitian anomnidimensionaldynamicconvolutionalnetworkforsingleimagesuperresolutiontasks AT xichen omnidimensionaldynamicconvolutionalnetworkforsingleimagesuperresolutiontasks AT ziangwu omnidimensionaldynamicconvolutionalnetworkforsingleimagesuperresolutiontasks AT weipingzhang omnidimensionaldynamicconvolutionalnetworkforsingleimagesuperresolutiontasks AT tingtingbi omnidimensionaldynamicconvolutionalnetworkforsingleimagesuperresolutiontasks AT chunweitian omnidimensionaldynamicconvolutionalnetworkforsingleimagesuperresolutiontasks |