Can anyone explain how to calculate AUC_DS, AUC_BW, and AUC_0 mentioned in this image?

Asked Mar 31 '22 at 09:47

Active Mar 31 '22 at 09:47

Viewed 35 times

I know AUC is for binary classification, but in a paper, the authors seemingly used AUC_DS, AUC_BW, and AUC_0 to compare vector values, like

ground truth of one sample [3,2,1,0,1,2,3]

prediction value of the sample [8.1, 7.9, 7.3, 6.9, 7.5, 7.8, 8.6]

they compared ground truth values and prediction values for many samples and gave a score. Can anyone explain how to calculate AUC_DS, AUC_BW, and AUC_0? Thank you.

asked Mar 31 '22 at 09:47

J S

Can you tell us which paper this is ? – Stratos supports the strike Mar 31 '22 at 10:07
The title is "Encoding Visual Sensitivity by MaxPol Convolution Filters for Image Sharpness Assessment" The preprint can be found here https://arxiv.org/pdf/1808.00617.pdf – J S Mar 31 '22 at 10:49

Can anyone explain how to calculate AUC_DS, AUC_BW, and AUC_0 mentioned in this image?

0 Answers0