· 2016
Learning Deep Features for Discriminative Localization
- 10,855citations
- 2016year
Short summary
A simple global average pooling layer, when trained on image-level labels, enables CNNs to localize objects with 37.1% top-5 error on ILSVRC 2014, without bounding box annotations.
AI-generated from the title and abstract; the full text is not read.
Key points
- Global average pooling layer enables CNNs for object localization using only image-level labels.
- Achieved 37.1% top-5 error for object localization on ILSVRC 2014.
- No bounding box annotations were used during training.
- Demonstrates CNNs' implicit attention on discriminative image regions.
AI-generated from the title and abstract; the full text is not read.
Abstract
In this work, we revisit the global average pooling layer proposed in [13], and shed light on how it explicitly enables the convolutional neural network (CNN) to have remarkable localization ability despite being trained on imagelevel labels. While this technique was previously proposed as a means for regularizing training, we find that it actually builds a generic localizable deep representation that exposes the implicit attention of CNNs on an image. Despite the apparent simplicity of global average pooling, we are able to achieve 37.1% top-5 error for object localization on ILSVRC 2014 without training on any bounding box annotation. We demonstrate in a variety of experiments that our network is able to localize the discriminative image regions despite just being trained for solving classification task1.
The authors' abstract, as published at the source., 2016 · DOI ↗
Continue with a free account
Ask the paper: 3 free questions a day about this paper; save it, get its citation, new summaries every day for your field. Takeaways are Premium.
Continue free on the webSign in with Google or Apple; no card needed. You come back to this paper.
On your phone:
Field: Computer Vision and Pattern Recognition
Computer Vision and Pattern RecognitionComputer Science