Latent Space Regularization for Unsupervised Domain Adaptation in Semantic Segmentation

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.

Automated summary

This paper introduces feature-level space-shaping regularization strategies to reduce domain discrepancy in semantic segmentation, achieving state-of-the-art results in the autonomous driving setting. By enforcing clustering objectives, perpendicularity constraints, and norm alignment goals on feature vectors corresponding to source and target samples, along with proposing a novel measure to capture the relative efficacy of adaptation strategies compared to supervised training, the effectiveness of such methods has been verified in multiple synthetic-to-real road scenes benchmarks.