-
Notifications
You must be signed in to change notification settings - Fork 36
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 reproduce Fig.3 Qualitative result about similarity map? #32
Comments
Hey @euncheolChoi, thank you for your interest in our work. Figure 3 of our paper compares the similarity of the facets across multiple layers. We use the dino_v2_sim_facets.py file for generating a joblib dump (to save the similarity maps). This save is done here AnyLoc/scripts/dino_v2_sim_facets.py Lines 358 to 366 in da93524
You can see the arguments of the script for more information. We use a custom script like facet_sim_visualization.py to finally get the figure. The script you gave seems to do dimensionality reduction using PCA and comparing similarity of a query point with all other points in the reduced space. What we do (instead) is that we take full-dimension features from a particular layer and facet and visualize similarity with a query feature (related to a point). |
Hello. Thank you for sharing your great work!
I've been trying to visualize a similarity map like the image in Fig. 3 from Anyloc paper for a few days now. However, I have not been able to get the right result.
The code below is the code I was working with. I am performing PCA on the norm_patchtoken obtained from dino, and generating a similarity map on the feature map obtained. However, the result is shown below.
Anyloc's repository doesn't seem to include the code to generate this similarity map. I was wondering if you could provide the code to reproduce the visualization result in Fig.3, or tell me how to visualize it.
Thank you.
Sample image
Feature map after PCA, Interpolation
Similarity map
The text was updated successfully, but these errors were encountered: