Wide receptive field networks for single image super-resolution

Yang, Haoran; Tong, Jiahui; Dou, Qingyu; Xiao, Long; Jeon, Gwanggil; Yang, Xiaobo

doi:10.1007/s11042-021-11258-4

Cited by 3 publications

(1 citation statement)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Compared with many traditional SISR methods based on machine learning, the simple structure of the SRCNN model shows remarkable performance in image super-resolution problems. Then, a large number of CNN-based models were proposed to obtain more accurate SISR results using different techniques to improve the quality of the reconstructed image: the design of the network structure with residuals [18][19][20][21][22]; generative adversarial networks [23]; neural architecture search [24,25]; various attention mechanisms [26], and other technologies [27,28]. With the improvement of architecture, this field has indeed made rich progress.…”

Section: Introductionmentioning

confidence: 99%

GRAN: Ghost Residual Attention Network for Single Image Super Resolution

Niu¹,

Wang²,

Zhu³

et al. 2023

Preprint

View full text Add to dashboard Cite

Recently, many works have designed wider and deeper networks to achieve higher image super-resolution performance. Despite their outstanding performance, they still suffer from high computational resources, preventing them from directly applying to embedded devices. To reduce the computation resources and maintain performance, we propose a novel Ghost Residual Attention Network (GRAN) for efficient super-resolution. This paper introduces Ghost Residual Attention Block (GRAB) groups to overcome the drawbacks of the standard convolutional operation, i.e., redundancy of the intermediate feature. GRAB consists of the Ghost Module and Channel and Spatial Attention Module (CSAM) to alleviate the generation of redundant features. Specifically, Ghost Module can reveal information underlying intrinsic features by employing linear operations to replace the standard convolutions. Reducing redundant features by the Ghost Module, our model decreases memory and computing resource requirements in the network. The CSAM pays more comprehensive attention to where and what the feature extraction is, which is critical to recovering the image details. Experiments conducted on the benchmark datasets demonstrate the superior performance of our method in both qualitative and quantitative. Compared to the baseline models,

show abstract