Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

How to generate the figure 4 of the paper? #5

Open
avijit9 opened this issue Jan 16, 2020 · 3 comments
Open

How to generate the figure 4 of the paper? #5

avijit9 opened this issue Jan 16, 2020 · 3 comments

Comments

@avijit9
Copy link

avijit9 commented Jan 16, 2020

Hi,

Is it possible to provide us with the code for reproducing the figure 4 from the paper?

@naraysa
Copy link
Owner

naraysa commented Jan 19, 2020

Unfortunately, I don't have access to that particular code anymore.

@avijit9
Copy link
Author

avijit9 commented Jan 19, 2020

Can you give some hints on this, please?

@naraysa
Copy link
Owner

naraysa commented Jan 21, 2020

We did the same steps as in paper (and detecionMAP code) to get the detections.

Network output of tcam for a category -> find threshold from min and max values of that tcam -> Obtain the continuous segments as valid detections. Only the valid detections part are shown in Fig 4. An additional step of normalization (softmax over class labels for each time step) is used for the purpose of visualization (scale issues) only.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants