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 draw Figure 9 and Figure 10 in your paper #6

Open
IceStreams opened this issue Sep 26, 2021 · 3 comments
Open

How to draw Figure 9 and Figure 10 in your paper #6

IceStreams opened this issue Sep 26, 2021 · 3 comments

Comments

@IceStreams
Copy link

Hello, thank you very much for your hard work, I think it is very meaningful.
How do you draw Figure 9 (Visualization of graphs on different data sets) and Figure 10(a) (Visualization of feature fusion) in your paper. I think this representation is very intuitive.Could you share it, thank you very much!

@qichaoliu
Copy link
Owner

Thanks.
How to draw Fig. 9:

  1. Compute spatial coordinate (x, y) of each superpixel by averaging all the coordinates of pixels in this superpixel. The average coordinate is taken as the position of its corresponding vertex (graph node) in figure.
  2. Save all the edge weights and colour them in figure according to the coordinates of nodes computed in (1).

How to draw Fig.10:

  1. Save the feature maps produced by CNN and GCN branches.
  2. Use a classifier like logistic regression to classify these feature maps in (1) according to training labels. Then visualize the predicted probabilities generated by this classifier.

@IceStreams
Copy link
Author

Thanks for your reply, about the fig9, how to colour the edge weights in figure? What method did you use to draw it? Is it the matplotlib library?

@qichaoliu
Copy link
Owner

Yes, matplotlib. If there is an edge between two nodes (coordinates), just draw a line between them, and set its colour!

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