Add DRAEM Model

[djdameln]

Description

• Adds model implementation of DRAEM.

• Fixes Add DRAEM Model #339

Known issues:

• Benchmarking results are not yet available. This algorithm is pretty slow (a few hours per training run), so it will take some time before I manage to collect the numbers for all categories.
• I'm using borrowed code for the Perlin noise generation. This module is a bit messy and will need to be refactored in the future.

Questions/Discussion:

• I feel that the augmenter class in anomalib/models/draem/augmenter.py could be useful for other models as well. We could consider moving it to components
• I ran into some problems with the pre-processor, because it assumes that we always want to normalize the input images to the ImageNet stats. For DRAEM this was not the case. It would be good to revisit this.

Changes

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• This change requires a documentation update

Checklist

• My code follows the pre-commit style and check guidelines of this project.
• I have performed a self-review of my code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• My changes generate no new warnings
• I have added tests that prove my fix is effective or that my feature works
• New and existing tests pass locally with my changes

[ashwinvaidya17]

Thanks! I have gone through my first round of review.

[samet-akcay]

Thanks!

[ashwinvaidya17]

I have a few minor comments

[ashwinvaidya17]

Thanks for the changes!

[samet-akcay]

We need to update
https://github.com/openvinotoolkit/anomalib/blob/development/third-party-programs.txt file

[hihunjin]

@samet-akcay Hi. Thanks for your implementation. Did you experiment this model on the whole MVTec AD datasets? If so, please share them.