Skip to content
Discussion options

You must be logged in to vote

Hello.
Image_ROC is score per image. This means that entire image is treated as one object, so score is assigned to it, which is then thresholded and compared to it's label (which is usually either 1 - nominal or 0 - anomalous).

Pixel_ROC is score per every pixel of the image. This means that we do similar comparison as in the case above, but each pixel is treated as one object. This means that it is thresholded and then compared to label, which is in this case a pixel in mask (again usually 0 where underlying pixel is good and 1 if it's anomalous). This is then applied over all pixels over all images.

Hope this helps 😃

Replies: 1 comment

Comment options

You must be logged in to vote
0 replies
Answer selected by samet-akcay
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Category
Q&A
Labels
None yet
2 participants
Converted from issue

This discussion was converted from issue #1150 on July 21, 2023 20:44.