Revisiting CycleGAN for semi-supervised segmentation

TL;DR

This paper introduces a semi-supervised segmentation method leveraging CycleGAN's style transfer and cycle consistency to improve performance with limited annotated data, outperforming recent approaches.

Contribution

It proposes a novel semi-supervised segmentation approach using cycle consistency in CycleGAN to enhance learning from unpaired images and masks.

Findings

Achieves 2-4% improvement over baseline in benchmarks.

Outperforms recent semi-supervised segmentation methods.

Effective especially in low labeled data scenarios.

Abstract

In this work, we study the problem of training deep networks for semantic image segmentation using only a fraction of annotated images, which may significantly reduce human annotation efforts. Particularly, we propose a strategy that exploits the unpaired image style transfer capabilities of CycleGAN in semi-supervised segmentation. Unlike recent works using adversarial learning for semi-supervised segmentation, we enforce cycle consistency to learn a bidirectional mapping between unpaired images and segmentation masks. This adds an unsupervised regularization effect that boosts the segmentation performance when annotated data is limited. Experiments on three different public segmentation benchmarks (PASCAL VOC 2012, Cityscapes and ACDC) demonstrate the effectiveness of the proposed method. The proposed model achieves 2-4% of improvement with respect to the baseline and outperforms recent approaches for this task, particularly in low labeled data regime.