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 evaluate on test data set? #10

Open
ricshaw opened this issue Jan 15, 2025 · 3 comments
Open

How to evaluate on test data set? #10

ricshaw opened this issue Jan 15, 2025 · 3 comments

Comments

@ricshaw
Copy link

ricshaw commented Jan 15, 2025

Hi, thanks for the great work.
In Table 1., how do you evaluate the novel view synthesis performance when you don't have appearance embedding for test images?

@yuehaowang
Copy link
Owner

Hi! Thanks for your question!

Appearance embeddings of testing views are set to all zeros as suggested in MipNeRF-360 & ZipNeRF.

@ricshaw
Copy link
Author

ricshaw commented Jan 16, 2025

You mean you don’t apply any bilateral filter at test time?

@yuehaowang
Copy link
Owner

Yes, in our method, no bilateral filter or only identity bilateral filters are applied to the test views.

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