SS-ADA: A Semi-Supervised Active Domain Adaptation Framework for Semantic Segmentation

TL;DR

SS-ADA is a semi-supervised active domain adaptation framework for semantic segmentation that intelligently selects unlabeled images for annotation, achieving high accuracy with limited labeled data in new driving scenarios.

Contribution

The paper introduces a novel SS-ADA framework combining active learning with semi-supervised segmentation, including an image-level acquisition strategy and IoU-based class weighting.

Findings

Achieves or surpasses supervised accuracy with only 25% labeled data.

Effective in synthetic-to-real and real-to-real domain adaptation.

Outperforms existing semi-supervised methods in accuracy and efficiency.

Abstract

Semantic segmentation plays an important role in intelligent vehicles, providing pixel-level semantic information about the environment. However, the labeling budget is expensive and time-consuming when semantic segmentation model is applied to new driving scenarios. To reduce the costs, semi-supervised semantic segmentation methods have been proposed to leverage large quantities of unlabeled images. Despite this, their performance still falls short of the accuracy required for practical applications, which is typically achieved by supervised learning. A significant shortcoming is that they typically select unlabeled images for annotation randomly, neglecting the assessment of sample value for model training. In this paper, we propose a novel semi-supervised active domain adaptation (SS-ADA) framework for semantic segmentation that employs an image-level acquisition strategy. SS-ADA integrates active learning into semi-supervised semantic segmentation to achieve the accuracy of supervised learning with a limited amount of labeled data from the target domain. Additionally, we design an IoU-based class weighting strategy to alleviate the class imbalance problem using annotations from active learning. We conducted extensive experiments on synthetic-to-real and real-to-real domain adaptation settings. The results demonstrate the effectiveness of our method. SS-ADA can achieve or even surpass the accuracy of its supervised learning counterpart with only 25% of the target labeled data when using a real-time segmentation model. The code for SS-ADA is available at https://github.com/ywher/SS-ADA.