Semi-supervised semantic segmentation needs strong, varied perturbations
This addresses the problem of improving semi-supervised learning for semantic segmentation, which is incremental as it builds on existing augmentation techniques.
The paper tackles the challenge of semi-supervised semantic segmentation by identifying that the data distribution lacks low-density regions between classes, making it difficult, and shows that using adapted CutOut and CutMix augmentations achieves state-of-the-art results on standard datasets.
Consistency regularization describes a class of approaches that have yielded ground breaking results in semi-supervised classification problems. Prior work has established the cluster assumption - under which the data distribution consists of uniform class clusters of samples separated by low density regions - as important to its success. We analyze the problem of semantic segmentation and find that its' distribution does not exhibit low density regions separating classes and offer this as an explanation for why semi-supervised segmentation is a challenging problem, with only a few reports of success. We then identify choice of augmentation as key to obtaining reliable performance without such low-density regions. We find that adapted variants of the recently proposed CutOut and CutMix augmentation techniques yield state-of-the-art semi-supervised semantic segmentation results in standard datasets. Furthermore, given its challenging nature we propose that semantic segmentation acts as an effective acid test for evaluating semi-supervised regularizers. Implementation at: https://github.com/Britefury/cutmix-semisup-seg.