Spatial Channel Attention for Deep Convolutional Neural Networks

Spatial Channel Attention for Deep Convolutional Neural Networks

Recently, the attention mechanism combining spatial and channel information has been widely used in various deep convolutional neural networks (CNNs), proving its great potential in improving model performance. However, this usually uses 2D global pooling operations to compress spatial information o...

Full description

Bibliographic Details
Main Authors: Tonglai Liu, Ronghai Luo, Longqin Xu, Dachun Feng, Liang Cao, Shuangyin Liu, Jianjun Guo
Format: Article
Language:English
Published: MDPI AG 2022-05-01
Series:Mathematics
Subjects:
attention mechanism
image classification
deep learning
cross-dimensional interaction
Online Access:https://www.mdpi.com/2227-7390/10/10/1750

Internet

https://www.mdpi.com/2227-7390/10/10/1750

Similar Items