Latent Space Regularization for Unsupervised Domain Adaptation in Semantic Segmentation
| Autor: | Marco Toldo, Pietro Zanuttigh, Umberto Michieli, Francesco Barbato |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
FOS: Computer and information sciences
business.industry Computer science Feature vector Computer Vision and Pattern Recognition (cs.CV) Computer Science - Computer Vision and Pattern Recognition Pattern recognition Semantics Regularization (mathematics) Convolutional neural network Domain (software engineering) Norm (mathematics) Segmentation Artificial intelligence Cluster analysis business |
| Zdroj: | CVPR Workshops |
| DOI: | 10.48550/arxiv.2104.02633 |
| Popis: | 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. Comment: Accepted at CVPR-WAD 2021, 11 pages, 7 figures, 1 tables |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|