Multi-Target Tracking with Trajectory Prediction and Re-Identification | |
Li Xuesong; Liu Yating; Wang Kunfeng; Yan Yong; Wang Fei-Yue | |
2020-02 | |
会议日期 | 22-24 Nov. 2019 |
会议地点 | Hangzhou, China |
关键词 | Multi-target tracking trajectory prediction reidentification deep learning computer vision |
DOI | 10.1109/CAC48633.2019.8996811 |
英文摘要 | Due to the complexity and clutter of real-world scenes, occlusion becomes a long-lasting difficulty in object tracking. Most existing tracking methods cannot effectively handle occlusion. In this paper, we propose a novel tracking framework that combines trajectory prediction and multi-cue appearance modeling to deal with the occlusion difficulty. When a target is completely occluded by background or other targets, it is unable to observe the target position. Therefore, we propose a Long Short-Term Memory (LSTM) model that merges attention mechanism and interaction module to predict the locations of all targets in the next frame. Considering that partial occlusion and inaccuracy of object bounding boxes often take place, we propose a multi-branch deep network architecture combining global and local features to realize accurate tracking and person re-identification (ReID). According to the experimental results on multiple benchmark datasets, our method achieves state-ofthe-art performance and outperforms many existing approaches. |
会议录出版者 | IEEE |
语种 | 英语 |
内容类型 | 会议论文 |
源URL | [http://ir.ia.ac.cn/handle/173211/39060] |
专题 | 自动化研究所_复杂系统管理与控制国家重点实验室_先进控制与自动化团队 |
作者单位 | 1.The State Key Laboratory for Management and Control of Complex Systems, Institute of Automation, Chinese Academy of Sciences 2.University of Chinese Academy of Sciences |
推荐引用方式 GB/T 7714 | Li Xuesong,Liu Yating,Wang Kunfeng,et al. Multi-Target Tracking with Trajectory Prediction and Re-Identification[C]. 见:. Hangzhou, China. 22-24 Nov. 2019. |
个性服务 |
查看访问统计 |
相关权益政策 |
暂无数据 |
收藏/分享 |
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