Learning to Segment from Scribbles using Multi-scale Adversarial Attention Gates

TL;DR

This paper introduces a novel weakly-supervised segmentation method using scribbles, multi-scale GANs, and adversarial attention gates, achieving performance comparable to fully-supervised models in medical and non-medical datasets.

Contribution

It proposes a multi-scale adversarial framework with attention gating conditioned by adversarial signals, improving object localization from scribbles in segmentation tasks.

Findings

Achieves segmentation performance comparable to fully-supervised models.

Effective in medical and non-medical datasets.

Extensible to semi-supervised, multi-source, and multi-task learning scenarios.

Abstract

Large, fine-grained image segmentation datasets, annotated at pixel-level, are difficult to obtain, particularly in medical imaging, where annotations also require expert knowledge. Weakly-supervised learning can train models by relying on weaker forms of annotation, such as scribbles. Here, we learn to segment using scribble annotations in an adversarial game. With unpaired segmentation masks, we train a multi-scale GAN to generate realistic segmentation masks at multiple resolutions, while we use scribbles to learn their correct position in the image. Central to the model's success is a novel attention gating mechanism, which we condition with adversarial signals to act as a shape prior, resulting in better object localization at multiple scales. Subject to adversarial conditioning, the segmentor learns attention maps that are semantic, suppress the noisy activations outside the objects, and reduce the vanishing gradient problem in the deeper layers of the segmentor. We evaluated our model on several medical (ACDC, LVSC, CHAOS) and non-medical (PPSS) datasets, and we report performance levels matching those achieved by models trained with fully annotated segmentation masks. We also demonstrate extensions in a variety of settings: semi-supervised learning; combining multiple scribble sources (a crowdsourcing scenario) and multi-task learning (combining scribble and mask supervision). We release expert-made scribble annotations for the ACDC dataset, and the code used for the experiments, at https://vios-s.github.io/multiscale-adversarial-attention-gates