arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2205.12840cs.CV
收藏

SALAD:用于分类、分割和检测的无源主动无标签域适应

SALAD: Source-free Active Label-Agnostic Domain Adaptation for Classification, Segmentation and Detection

University of Maryland College Park(马里兰大学学院公园分校);Adobe Research(奥多比研究院)

Divya Kothandaraman, Sumit Shekhar, Abhilasha Sancheti, Manoj Ghuhan, Tripti Shukla, Dinesh Manocha

AI总结:

针对源数据不可用、目标域标注预算少且标签空间偏移的域适配问题,提出SALAD方法,通过引导注意力迁移网络和主动学习函数联合优化样本选择与知识迁移,在三类视觉任务上较依赖源数据的方法提升0.5%-31.3%。

摘要:

我们提出了一种名为SALAD的新方法,用于解决具有挑战性的视觉任务:将预训练的“源”域网络适配到“目标”域,该任务在目标域中只有少量标注预算,且标签空间存在偏移。此外,由于隐私问题或其他原因,该任务假设源数据无法用于适配。我们假设此类系统需要联合优化两个任务:(i)从目标域中选择固定数量的样本进行标注;(ii)将预训练网络的知识迁移到目标域。为此,SALAD包含一个新颖的引导注意力迁移网络(GATN)和一个主动学习函数HAL。GATN实现了从预训练网络到目标网络的特征蒸馏,并辅以HAL利用迁移能力和不确定性标准挖掘的目标样本。SALAD有三个关键优势:(i)它与任务无关,可应用于分类、分割和检测等各种视觉任务;(ii)它可以处理从预训练源网络到目标域的输出标签空间偏移;(iii)它不需要访问源数据进行适配。我们在3个视觉任务上进行了广泛实验,即数字分类(MNIST、SVHN、VISDA)、合成(GTA5)到真实(CityScapes)图像分割,以及文档布局检测(PubLayNet到DSSE)。我们表明,与假设可访问大量带标注源数据进行适配的现有适配方法相比,我们的无源方法SALAD在不同数据集和任务上的性能提升了0.5%-31.3%。

英文摘要:

We present a novel method, SALAD, for the challenging vision task of adapting a pre-trained "source" domain network to a "target" domain, with a small budget for annotation in the "target" domain and a shift in the label space. Further, the task assumes that the source data is not available for adaptation, due to privacy concerns or otherwise. We postulate that such systems need to jointly optimize the dual task of (i) selecting fixed number of samples from the target domain for annotation and (ii) transfer of knowledge from the pre-trained network to the target domain. To do this, SALAD consists of a novel Guided Attention Transfer Network (GATN) and an active learning function, HAL. The GATN enables feature distillation from pre-trained network to the target network, complemented with the target samples mined by HAL using transfer-ability and uncertainty criteria. SALAD has three key benefits: (i) it is task-agnostic, and can be applied across various visual tasks such as classification, segmentation and detection; (ii) it can handle shifts in output label space from the pre-trained source network to the target domain; (iii) it does not require access to source data for adaptation. We conduct extensive experiments across 3 visual tasks, viz. digits classification (MNIST, SVHN, VISDA), synthetic (GTA5) to real (CityScapes) image segmentation, and document layout detection (PubLayNet to DSSE). We show that our source-free approach, SALAD, results in an improvement of 0.5%-31.3%(across datasets and tasks) over prior adaptation methods that assume access to large amounts of annotated source data for adaptation.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑