A Simple and Light-Weight Attention Module for Convolutional Neural Networks

Many aspects of deep neural networks, such as depth, width, or cardinality, have been studied to strengthen the representational power. In this work, we study the effect of attention in convolutional neural networks and present our idea in a simple self-contained module, called Bottleneck Attention...

Full description

Saved in:
Bibliographic Details
Published in:International journal of computer vision 2020-04, Vol.128 (4), p.783-798
Main Authors: Park, Jongchan, Woo, Sanghyun, Lee, Joon-Young, Kweon, In So
Format: Article
Language:English
Subjects:
Ablation
Artificial Intelligence
Artificial neural networks
Computer Imaging
Computer Science
Feature maps
Image Processing and Computer Vision
Modules
Neural networks
Pattern Recognition
Pattern Recognition and Graphics
Special Issue on Machine Vision with Deep Learning
Vision
Weight reduction
Wireless networks
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Many aspects of deep neural networks, such as depth, width, or cardinality, have been studied to strengthen the representational power. In this work, we study the effect of attention in convolutional neural networks and present our idea in a simple self-contained module, called Bottleneck Attention Module (BAM). Given an intermediate feature map, BAM efficiently produces the attention map along two factorized axes, channel and spatial , with negligible overheads. BAM is placed at bottlenecks of various models where the downsampling of feature maps occurs, and is jointly trained in an end-to-end manner. Ablation studies and extensive experiments are conducted in CIFAR-100/ImageNet classification, VOC2007/MS-COCO detection, super resolution and scene parsing with various architectures including mobile-oriented networks. BAM shows consistent improvements over all experiments, demonstrating the wide applicability of BAM. The code and models are available at https://github.com/Jongchan/attentionmodule .
ISSN:0920-5691
1573-1405
DOI:10.1007/s11263-019-01283-0