Skip to content
This repository has been archived by the owner on Feb 20, 2020. It is now read-only.

Visualizing where the Convolution Network is looking through CAM.

License

Notifications You must be signed in to change notification settings

alokwhitewolf/Class-Activation-Mappings

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

9 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Class-Activation-Mappings

Chainer implementation of Learning Deep Features for Discriminative Localization

This implementation uses ResNet50 architecture to infer the class activations. Image

Class activation maps are a simple technique to get the discriminative image regions used by a CNN to identify a specific class in the image. In other words, a class activation map (CAM) lets us see which regions in the image were relevant to this class. The authors of the paper show that this also allows re-using classifiers for getting good localization results, even when training without bounding box coordinates data. This also shows how deep learning networks already have some kind of a built in attention mechanism.This should be useful for debugging the decision process in classification networks.

Image

Dependencies

To run

Trained Chainer model for ResNet50 is stored in model. To test,

python main.py -i <path-to-image>