MSANet: Multi-scale attention networks for image classification

Cao, Ping; Xie, Fangxin; Zhang, Shichao; Zhang, Zuping; Zhang, Jianfeng

doi:10.1007/s11042-022-12792-5

Cited by 4 publications

(1 citation statement)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is worthy to note that the approach of combining multi-scale features using attention has been commonly used in other vision problems such as classification, 7 segmentation, 8 , 9 enhancement, 10 and inpainting 11 . However, they obtain different scales of local attention by controlling the size of the convolution kernel, using dilated convolutions with different dilation rates, or using different sizes of feature maps.…”

Section: Introductionmentioning

confidence: 99%

Image super-resolution using multi-scale non-local attention

Kim

Park

2023

J. Electron. Imag.

View full text Add to dashboard Cite

Recent convolutional neural network (CNN)-based super-resolution (SR) studies incorporated non-local attention (NLA), to consider the long-range feature correlations and achieve considerable performance improvement. Here, we proposed an innovative NLA scheme called multi-scale NLA (MS-NLA) that computes NLAs at multiple scales and fuses them. To effectively fuse NLAs, we also proposed two learning-based methods and analyzed their performance on a recurrent SR network. Furthermore, the effect of weight sharing in the fusion methods is analyzed as well. In 2× and 4× SR experiments on benchmark datasets, our method had higher PSNR values of 0.295 and 0.148 dB on average than those using single-scale NLA and crossscale NLA, respectively, and produced visually more pleasing SR results. The weight sharing had a limited but positive effect, depending on datasets. The source code is accessible at https://github .com/Dae12-Han/MSNLN.

show abstract