Multiple Source Domain Adaptation for Multiple Object Tracking in Satellite Video
文献类型:期刊论文
作者 | Zheng, Xiangtao3; Cui, Haowen1,2; Lu, Xiaoqiang3 |
刊名 | IEEE Transactions on Geoscience and Remote Sensing |
出版日期 | 2023 |
卷号 | 61页码:1-11 |
ISSN号 | 01962892;15580644 |
关键词 | Cross-domain recognition deep neural networks multiple object tracking (MOT) object detection satellite video |
DOI | 10.1109/TGRS.2023.3336665 |
产权排序 | 2 |
英文摘要 | Satellite videos capture the dynamic changes in a large observed sense, which provides an opportunity to track the object trajectories. However, existing multiple object tracking (MOT) methods require massive video annotations, which is time-consuming and fallible. To alleviate this problem, this article proposes a cross-domain multiple object tracker (CDTrack) to learn knowledge from multiple source domains. First, a cross-domain object detector with multilevel domain alignment is constructed to learn domain-invariant knowledge between remote sensing images and satellite videos. Second, the proposed method adopts a bidirectional teacher-student framework to fuse multiple source domains. Two teacher-student models learn different domain knowledge and teach mutually each other. With mutual learning, the proposed method alleviates the discrepancies between different domains. Finally, a simple weakly supervised Re-IDentification (Re-ID) model is proposed for long-term association. Experimental results on the satellite video datasets demonstrate that the proposed method can achieve great performance without satellite video annotations. The code is available at https://github.com/XiangtaoZheng/CDTrack. © 1980-2012 IEEE. |
语种 | 英语 |
出版者 | Institute of Electrical and Electronics Engineers Inc. |
源URL | [http://ir.opt.ac.cn/handle/181661/97054] |
专题 | 西安光学精密机械研究所_光学影像学习与分析中心 |
通讯作者 | Zheng, Xiangtao |
作者单位 | 1.University of Chinese Academy of Sciences, Beijing; 100049, China 2.Xi'an Institute of Optics and Precision Mechanics, Chinese Academy of Sciences, Key Laboratory of Spectral Imaging Technology Cas, Xi'an; 710119, China; 3.Fuzhou University, College of Physics and Information Engineering, Fuzhou; 350108, China; |
推荐引用方式 GB/T 7714 | Zheng, Xiangtao,Cui, Haowen,Lu, Xiaoqiang. Multiple Source Domain Adaptation for Multiple Object Tracking in Satellite Video[J]. IEEE Transactions on Geoscience and Remote Sensing,2023,61:1-11. |
APA | Zheng, Xiangtao,Cui, Haowen,&Lu, Xiaoqiang.(2023).Multiple Source Domain Adaptation for Multiple Object Tracking in Satellite Video.IEEE Transactions on Geoscience and Remote Sensing,61,1-11. |
MLA | Zheng, Xiangtao,et al."Multiple Source Domain Adaptation for Multiple Object Tracking in Satellite Video".IEEE Transactions on Geoscience and Remote Sensing 61(2023):1-11. |
入库方式: OAI收割
来源:西安光学精密机械研究所
其他版本
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。