Grad_CAM_plus_plus
Grad_CAM_plus_plus copied to clipboard
adityac94
A generalized gradient-based CNN visualization technique
Dear @adityac94 , @tataiani If I understand correctly, you use the following assumption to go from eq.7 to eq.8, and from eq.8 to eq.9: $\frac{\partial^2 Y^C}{\partial A^k_{ab} \partial A^k_{ij}} =...
Hello, I am having a hard time getting equation (8) from (7). Note that in eq. (7) index k is a summation index, hence it is better to rename it...
I noticed that equation (19) in the paper:  can be simplified by dividing numerator and denominator by the squared gradient:  The simplified version (19") is more efficient (no...
In your paper, you give a evaluation indexes: average drop and average increase. However, I didn't see these in your code.
In your paper, you give a evaluation indexes: average drop and average increase. We can see a table about these indexes in your paper, so I have a qusetion about...
I ran through your code and found that the map is dark. How can it become a heat map?thanks! @adityac94
How can I remove the blue color in the heatmap background because in most images it is filling them.
The google drive sharing link given in the README refers to an empty folder. Thus, the file vgg16.npy could not be found. This appears to me as an urgent issue...
Hi. In the paper you derived equations for the case when softmax function is used to obtain final class probabilities. (Equations 21-24) I am unable to find an implementation of...
Hi, it seems to me that you can simplify the equation Equ 17 by dividing both numerator and denominator by (dS/dA)^2, to get : > alpha = 1/(2 + sum(A_ab)*(dS/dA))...
