Colour augmentation for improved semi-supervised semantic segmentation

TL;DR

This paper introduces colour augmentation to enhance semi-supervised semantic segmentation by preventing reliance on colour shortcuts, leading to improved performance on complex photographic images.

Contribution

It proposes using colour augmentation to address shortcut learning in semi-supervised segmentation, inspired by self-supervised learning insights.

Findings

Colour augmentation improves segmentation accuracy.

Prevents reliance on colour shortcuts.

Effective on challenging photographic datasets.

Abstract

Consistency regularization describes a class of approaches that have yielded state-of-the-art results for semi-supervised classification. While semi-supervised semantic segmentation proved to be more challenging, a number of successful approaches have been recently proposed. Recent work explored the challenges involved in using consistency regularization for segmentation problems. In their self-supervised work Chen et al. found that colour augmentation prevents a classification network from using image colour statistics as a short-cut for self-supervised learning via instance discrimination. Drawing inspiration from this we find that a similar problem impedes semi-supervised semantic segmentation and offer colour augmentation as a solution, improving semi-supervised semantic segmentation performance on challenging photographic imagery.