In the field of domain adaptation, a trade-off exists between the model performance and the number of target domain annotations. Active learning, maximizing model performance with few informative labeled data, comes in handy for such a scenario. In this work, we present D2ADA, a general active domain adaptation framework for semantic segmentation. To adapt the model to the target domain with minimum queried labels, we propose acquiring labels of the samples with high probability density in the target domain yet with low probability density in the source domain, complementary to the existing source domain labeled data. To further facilitate labeling efficiency, we design a dynamic scheduling policy to adjust the labeling budgets between domain exploration and model uncertainty over time. Extensive experiments show that our method outperforms existing active learning and domain adaptation baselines on two benchmarks, GTA5 -> Cityscapes and SYNTHIA -> Cityscapes. With less than 5% target domain annotations, our method reaches comparable results with that of full supervision. Our code is publicly available at https://github.com/tsunghan-wu/D2ADA.
翻译:在领域适应领域,模型性能与目标域说明数量之间存在一种权衡。积极学习,尽量扩大模型性能,加上很少的信息标签数据,对于这种情景是有用的。在这项工作中,我们提出D2AD,这是用于语义分化的一般活跃域适应框架。为了使模型适应目标域,并贴上最小的查询标签,我们提议获取目标域中高概率密度样本的标签,但来源域的概率密度较低,以补充现有的源域域标记数据。为进一步促进标签效率,我们设计了动态时间安排政策,以调整域探索与模型不确定性之间的标签预算。广泛的实验表明,我们的方法优于现有两个基准(GTA5 - > 城市景观和 SYNTHIA - > 城市景区)的现有积极学习和域适应基线。在不到5%的目标域说明下,我们的方法与全面监督的结果相似。我们的代码可在https://github.com/tsunghan-wu/D2ADADADADA中公开查阅。