Menu

Top-down Neural Attention by Excitation Backprop

calendar icon Oct 24, 2016 1657 views
split view icon
video icon
presentation icon
video with chapters icon
video thumbnail
Pause
Mute
speed icon
speed icon
0.25
0.5
0.75
1
1.25
1.5
1.75
2

We aim to model the top-down attention of a Convolutional Neural Network (CNN) classifier for generating task-specific attention maps. Inspired by a top-down human visual attention model, we propose a new backpropagation scheme, called Excitation Backprop, to pass along top-down signals downwards in the network hierarchy via a probabilistic Winner-Take-All process. Furthermore, we introduce the concept of contrastive attention to make the top-down attention maps more discriminative. In experiments, we demonstrate the accuracy and generalizability of our method in weakly supervised localization tasks on the MS COCO, PASCAL VOC07 and ImageNet datasets. The usefulness of our method is further validated in the text-to-region association task. On the Flickr30k Entities dataset, we achieve promising performance in phrase localization by leveraging the top-down attention of a CNN model that has been trained on weakly labeled web images.

RELATED CATEGORIES

MORE VIDEOS FROM THE SAME CATEGORIES

Except where otherwise noted, content on this site is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International license.