The reviewed record of science sign in
Pith

arxiv: 2104.02633 · v3 · pith:EAYQCJD7 · submitted 2021-04-06 · cs.CV

Latent Space Regularization for Unsupervised Domain Adaptation in Semantic Segmentation

Reviewed by Pithpith:EAYQCJD7open to challenge →

classification cs.CV
keywords segmentationsemantictheyadaptationdatadomainregularizationtraining
0
0 comments X
read the original 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.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.