search

cvlab-yonsei / MNAD

An official implementation of "Learning Memory-guided Normality for Anomaly Detection" (CVPR 2020) in PyTorch.
334 stars 82 forks source link

Unable to save recreated images and run model real time on a video #8

Closed kevins99 closed 3 years ago

kevins99 commented 3 years ago

Hello! Thank you for your great contribution towards anomaly detection. I am currently getting started with your repo and have trained the model on the UCSD dataset. However, how do I create a demo showing the anomaly scores real time as the video plays and how do I generate the reconstructions. Is a script available for the same? Thank you!

RohDonghyeon commented 3 years ago

I am interested in your question. Have you implemented the appropriate method?