mala-lab / ADShift

Official PyTorch implementation of the ICCV'23 paper “Anomaly Detection under Distribution Shift”
25 stars 4 forks source link

What's your anomaly scores formulation??? In your paper, it didn't appear! please tell me, thx! #6

Open LingoRihood opened 8 months ago

caothientri2001vn commented 7 months ago

The anomaly scores have been described in the Appendix or you can find it in the paper "Anomaly Detection via Reverse Distillation from One-Class Embedding", I follow the same way of calculation.