Giters

openvinotoolkit / anomalib

An anomaly detection library comprising state-of-the-art algorithms and features such as experiment management, hyper-parameter optimization, and edge inference.

Home Page:https://anomalib.readthedocs.io/en/latest/

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

[Task]: Is there any plan to add Unsupervised Continual Anomaly Detection with Contrastively-learned Prompt?

fat-921 opened this issue · comments

commented

What is the motivation for this task?

1.The current anomaly detection can only train one model for one class. What if only one model is trained for multiple classes?
2.How to add continuous learning to anomaly detection model training?

Describe the solution you'd like

I saw a paper like Unsupervised Continual Anomaly Detection with Contrastively-learned Prompt in AAAI-2024 [paper link] [code link] , it's a great idea to work with, and it might be even better to add it to anomalib.

Additional context

No response

commented

This looks like an interesting approach. I'll add it to the feature request section with Good First Issue label in case someone wants to work on it.

Thanks for sharing it!