search

marco-rudolph / differnet

This is the official repository to the WACV 2021 paper "Same Same But DifferNet: Semi-Supervised Defect Detection with Normalizing Flows" by Marco Rudolph, Bastian Wandt and Bodo Rosenhahn.
217 stars 68 forks source link

Qualitative results between CS-Flow and DifferNet on MTD dataset #37

Closed sungwool closed 2 years ago

sungwool commented 2 years ago

Hello, I'm interested in your method.

While reading your papers(DifferNet, CS-Flow), I got a question.

The 1-NN performance in the DifferNet paper is recorded as 80.0 but in the CS-Flow paper, it is recorded as 97.7. (Method for "Are pretrained cnns good feature extractors for anomaly detection in surveillance videos?")

May I know the difference between the two results? And, Is there any difference in the experimental environment?

Thank you :)

marco-rudolph commented 2 years ago

Hello, first, the feature extractor has been changed in CS-Flow (EfficientNet-B5 instead of AlexNet). Second, in CS-Flow the nearest neighbor was calculated based on the full-sized feature map, while in DifferNet averaged feature maps, i.e. vectors, were used.

Best regards Marco