[edsykes]

I had a read through this and I couldn't really tell if there was something novel here?

I understand that perturbations and generating new examples from labelled examples is a pretty normal park of the process when you only have a limited number of examples available.

[amitness]

The novelty is in applying 2 perturbations to available unlabeled images and use them as part of training. This is different than what you are describing about applying augmentations to labeled images to increase data size.

[daenz]

My immediate question was "how do you use unlabeled images for training?" But then I decided to read the paper :) The answer is:

Two different perturbations to the same image should have the same predicted label by the model, even if it doesn't know what the correct label is. That information can be used in the training.