CSMOT: Make One-Shot Multi-Object Tracking in Crowded Scenes Great Again | |
Hou, Haoxiong1,2; Shen, Chao1,2; Zhang, Ximing1; Gao, Wei1 | |
作者部门 | 空间光学技术研究室 |
2023-04 | |
发表期刊 | SENSORS |
ISSN | 1424-8220 |
卷号 | 23期号:7 |
产权排序 | 1 |
摘要 | The current popular one-shot multi-object tracking (MOT) algorithms are dominated by the joint detection and embedding paradigm, which have high inference speeds and accuracy, but their tracking performance is unstable in crowded scenes. Not only does the detection branch have difficulty in obtaining the accurate object position, but the ambiguous appearance of features extracted by the re-identification (re-ID) branch also leads to identity switches. Focusing on the above problems, this paper proposes a more robust MOT algorithm, named CSMOT, based on FairMOT. First, on the basis of the encoder-decoder network, a coordinate attention module is designed to enhance the information interaction between channels (horizontal and vertical coordinates), which improves its object-detection abilities. Then, an angle-center loss that effectively maximizes intra-class similarity is proposed to optimize the re-ID branch, and the extracted re-ID features are made more discriminative. We further redesign the re-ID feature dimension to balance the detection and re-ID tasks. Finally, a simple and effective data association mechanism is introduced, which associates each detection instead of just the high-score detections during the tracking process. The experimental results show that our one-shot MOT algorithm achieves excellent tracking performance on multiple public datasets and can be effectively applied to crowded scenes. In particular, CSMOT decreases the number of ID switches by 11.8% and 33.8% on the MOT16 and MOT17 test datasets, respectively, compared to the baseline. |
关键词 | one-shot multi-object tracking re-ID coordinate attention angle-center loss data association |
DOI | 10.3390/s23073782 |
收录类别 | SCI |
语种 | 英语 |
WOS记录号 | WOS:000970227200001 |
出版者 | MDPI |
引用统计 | |
文献类型 | 期刊论文 |
条目标识符 | http://ir.opt.ac.cn/handle/181661/96442 |
专题 | 空间光学技术研究室 |
通讯作者 | Zhang, Ximing; Gao, Wei |
作者单位 | 1.Chinese Acad Sci, Xian Inst Opt & Precis Mech, Xian 710119, Peoples R China 2.Univ Chinese Acad Sci, Beijing 101408, Peoples R China |
推荐引用方式 GB/T 7714 | Hou, Haoxiong,Shen, Chao,Zhang, Ximing,et al. CSMOT: Make One-Shot Multi-Object Tracking in Crowded Scenes Great Again[J]. SENSORS,2023,23(7). |
APA | Hou, Haoxiong,Shen, Chao,Zhang, Ximing,&Gao, Wei.(2023).CSMOT: Make One-Shot Multi-Object Tracking in Crowded Scenes Great Again.SENSORS,23(7). |
MLA | Hou, Haoxiong,et al."CSMOT: Make One-Shot Multi-Object Tracking in Crowded Scenes Great Again".SENSORS 23.7(2023). |
条目包含的文件 | ||||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
CSMOT Make One-Shot (2806KB) | 期刊论文 | 出版稿 | 限制开放 | CC BY-NC-SA | 请求全文 |
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。
修改评论