Simonyan, Karen, Andrea Vedaldi, and Andrew Zisserman. “Deep inside convolutional networks: Visualising image classification models and saliency maps.” arXiv preprint arXiv:1312.6034 (2013). (Citations: 227).
1 Deconv Approach
Compute gradient of class k wrt the image using backpropagation.
1.1 Class Saliency Map
The class saliency map M ∈ R^D0 ×H0 ×W0 is
The white areas indicate that if you jitter the pixels of the object, they will influence the class score in some way.
1.2 Weakly Supervised Object Localiz