Jpcbertoldo/mvtec ad loco 2

[jpcbertoldo]

Description

Create a new dataset: MVTec LOCO Anomaly Detection.

"LOCO" stands for "LOgical COnstraints"

I based myself on anomalib/data/mvtec.py.

• Fixes Add support for MVTEC LOCO AD #471

imread_strategy

The dataset supports an option imread_strategy which allows the user how to choose when the images are loaded:

• onthefly: behaviour I found in mvtec.py, the images are loaded upon demand during the training;
• preload: all the images are cached in the memory (RAM, not GPU) when the dataset is being initialized.

anotype and super_anotype

Besides providing the binary label, I also create the dataset with two other categorical values:

• super_anotype: is it a logical or structural anomaly? (or a normal?)
• anotype: "what is the problem with the image?", mvtec ad also has different types of anomalies for each category but this is particularly more interesting here because there are many types of logical violations possible.

I specifically included this because I am interested in evaluating separately by those types but I will later create an issue for that feature.

mask vs. masks

MVTec LOCO's logical anomalies may include several anoamlies in a single image and to properly evaluate them one needs to consider them separately so they are segmented in different mask files in the ground truth.

Since the rest of library expects a tensor mask (SINGULAR), I merge them all into a single binary maks (with loss information because they cannot be separated anymore).

In order to later peform proper evaluation there is a second tensor masks (PLURAL) which encodes each anomalous region with a different value (0 is a normal pixel, and 1, 2, ..., N are anomalous pixels).

things in MVTecAD but not in MVTecLOCO

1) self.transform_config_val = self.transform_config_train

        if self.transform_config_train is not None and self.transform_config_val is None:
            self.transform_config_val = self.transform_config_train

Is there a good reason for assuming this?

For me it could make sense that self.transform_config_val could have light data augmentations (say, tiny brightness changes) but that should not be repeated in the validation set.

2) split_normal_images_in_train_set(samples, split_ratio, seed)

MVTec LOCO already defines fixed validation sets so i did not include the option of doing it dinamically like in MVTec AD.

---

Checklists

Changes

• Bug fix (non-breaking change which fixes an issue)
• Refactor (non-breaking change which refactors the code base)
• 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

[review-notebook-app]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[jpcbertoldo]

self.pre_process is being called for the 3rd time here, will that create any problems?

I'm thinking that maybe the random transforms will apply the same transform every two times (for the image and for the mask).

[jpcbertoldo]

The images in this dataset are not squared.
The ratio of widh/height can end up too different than the original image when the image size is given as an int.

Maybe we should add a warning here?

[jpcbertoldo]

ready to review

[jpcbertoldo]

Please ignore for now, I will wait for PR #558.
Also, I realize that the pre-loading of data is probably unnecessary so I'll remove it to make the PR simpler.

[djdameln]

@jpcbertoldo I am closing this because it has been inactive for a long time and is outdated. Feel free to re-open if you resume working on this.