当前位置: X-MOL 学术Mach. Vis. Appl. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Object detection based on semi-supervised domain adaptation for imbalanced domain resources
Machine Vision and Applications ( IF 2.4 ) Pub Date : 2020-03-25 , DOI: 10.1007/s00138-020-01068-3
Wei Li , Meng Wang , Hongbin Wang , Yafei Zhang

On specified scenarios, models trained on specific datasets (source domain) can generalize well to novel scenes (target domain) via knowledge transfer. However, these source detectors might not be perfectly aligned with a low target resource due to the imbalanced and inconsistent domain shift involved. In this paper, we propose a semi-supervised detector that adapts the domain shifts on both appearance and semantic levels. Based on this, two components are introduced as appearance adaptation networks with instance and batch normalization, and semantic adaptation networks where an adversarial transferring procedure is embedded by re-weighting the discriminator loss to improve the feature alignments between the two domains with imbalanced scales. Furthermore, a self-paced training procedure is performed to re-train the detector by alternately generating pseudo-labels in the target domain from easy to hard. In our experiments, an empirical analysis of the proposed framework is conducted by evaluating performance in various datasets such as Cityscapes and VOC0712, and the results verify the higher accuracy and effectiveness of the proposed detector in comparison with state-of-the-art detectors.

中文翻译:

基于半监督域自适应的不平衡域资源目标检测

在特定情况下,通过知识转移,在特定数据集(源域)上训练的模型可以很好地推广到新场景(目标域)。但是,由于涉及的不平衡且不一致的域偏移,这些源检测器可能无法与低目标资源完美对齐。在本文中,我们提出了一种半监督检测器,该检测器在外观和语义级别上都适应了域移位。在此基础上,引入了两个组件,分别是具有实例和批处理规范化的外观适应网络,以及语义适应网络,在语义适应网络中,通过对加权值进行重新加权来嵌入对抗性传递过程,以改善比例不平衡的两个域之间的特征对齐。此外,通过在目标域中从易到难交替生成伪标记,执行自定进度的训练过程以重新训练检测器。在我们的实验中,通过评估各种数据集(例如Cityscapes和VOC0712)的性能,对提出的框架进行了实证分析,结果证明,与最新的探测器相比,提出的探测器具有更高的准确性和有效性。
更新日期:2020-03-25
down
wechat
bug