A lightweight image super-resolution method based on a hybrid CNN-Transformer architecture
林承浩 1吴丽君1
扫码查看
点击上方二维码区域，可以放大扫码查看
作者信息
1. 福州大学 物理与信息工程学院,福建 福州 350108
折叠
摘要
针对基于混合构架的图像超分模型通常需要较高计算成本的问题,提出了一种基于CNN-Transformer混合构架的轻量图像超分网络STSR(Swin-Transformer-based Single Image Super-Resolution).首先,提出了一种并行特征提取的特征增强模块(Feature Enhancement Block,FEB),由卷积神经网络(Convolutional Neural Network,CNN)和轻量型Transformer网络并行地对输入图像进行特征提取,再将提取到的特征进行特征融合.其次,设计了一种动态调整模块(Dynamic Adjustment,DA),使得网络能根据输入图像来动态调整网络的输出,减少网络对无关信息的依赖.最后,采用基准数据集来测试网络的性能,实验结果表明STSR在降低模型参数量的前提下仍然保持较好的重建效果.
Abstract
In order to address the problem that image super-segmentation models based on hybrid architectures usually require high computational cost,this study proposes a lightweight image super-segmentation network STSR(Swin-Transformer-based Sin-gle Image Super-Resolution)based on a hybrid CNN-Transformer architecture.Firstly,this paper proposes a Feature Enhance-ment Block(FEB)for parallel feature extraction,which consists of a Convolutional Neural Network(CNN)and a lightweight Transformer Network to extract features from the input image in parallel,and then the extracted features are fused to the features.Secondly,this paper designs a Dynamic Adjustment(DA)module,which enables the network to dynamically adjust the output of the network according to the input image,reducing the network's dependence on irrelevant information.Finally,some benchmark datasets are used to test the performance of the network,and the experimental results show that STSR still maintains a better re-construction effect under the premise of reducing the number of model parameters.
维普
万方数据