Distance transform regression for spatially-aware deep semantic segmentation

TL;DR

This paper proposes a novel regularization method for semantic segmentation that uses distance transform regression, improving boundary accuracy and shape delineation with minimal network modifications.

Contribution

It introduces a multi-task learning approach combining classification and distance transform regression to enhance segmentation quality.

Findings

Significant boundary accuracy improvements across datasets.

Low overhead with minimal network modifications.

Effective with various architectures.

Abstract

Understanding visual scenes relies more and more on dense pixel-wise classification obtained via deep fully convolutional neural networks. However, due to the nature of the networks, predictions often suffer from blurry boundaries and ill-segmented shapes, fueling the need for post-processing. This work introduces a new semantic segmentation regularization based on the regression of a distance transform. After computing the distance transform on the label masks, we train a FCN in a multi-task setting in both discrete and continuous spaces by learning jointly classification and distance regression. This requires almost no modification of the network structure and adds a very low overhead to the training process. Learning to approximate the distance transform back-propagates spatial cues that implicitly regularizes the segmentation. We validate this technique with several architectures on various datasets, and we show significant improvements compared to competitive baselines.