Latent Space Regularization for Unsupervised Domain Adaptation in Semantic Segmentation

TL;DR

This paper introduces feature-level regularization techniques for unsupervised domain adaptation in semantic segmentation, improving generalization from synthetic to real data in autonomous driving scenarios.

Contribution

It proposes novel regularization strategies and a new measure to evaluate adaptation effectiveness, achieving state-of-the-art results in synthetic-to-real domain adaptation.

Findings

Improved domain adaptation performance in autonomous driving datasets.

Effective feature clustering and alignment strategies.

State-of-the-art results in multiple benchmarks.

Abstract

Deep convolutional neural networks for semantic segmentation achieve outstanding accuracy, however they also have a couple of major drawbacks: first, they do not generalize well to distributions slightly different from the one of the training data; second, they require a huge amount of labeled data for their optimization. In this paper, we introduce feature-level space-shaping regularization strategies to reduce the domain discrepancy in semantic segmentation. In particular, for this purpose we jointly enforce a clustering objective, a perpendicularity constraint and a norm alignment goal on the feature vectors corresponding to source and target samples. Additionally, we propose a novel measure able to capture the relative efficacy of an adaptation strategy compared to supervised training. We verify the effectiveness of such methods in the autonomous driving setting achieving state-of-the-art results in multiple synthetic-to-real road scenes benchmarks.