Visual vs internal attention mechanisms in deep neural networks for image classification and object detection

作者：

Highlights：

• Visual attention integration into convolutional neural networks.

• Attention mechanisms in classification and object detection tasks.

• Visual attention as subjective saliency maps allows faster convergence and better performance.

摘要

•Visual attention integration into convolutional neural networks.•Attention mechanisms in classification and object detection tasks.•Visual attention as subjective saliency maps allows faster convergence and better performance.

论文关键词：Deep learning,Image classification,Object detection,Visual attention,Saliency maps

论文评审过程：Received 20 June 2020, Revised 26 October 2021, Accepted 30 October 2021, Available online 3 November 2021, Version of Record 12 November 2021.

论文官网地址：https://doi.org/10.1016/j.patcog.2021.108411