arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1802.00977cs.CVcs.AI
收藏

Pose Flow:高效的在线姿势跟踪

Pose Flow: Efficient Online Pose Tracking

Shanghai Jiao Tong University(上海交通大学)

Yuliang Xiu, Jiefeng Li, Haoyu Wang, Yinghong Fang, Cewu Lu

摘要:

无约束视频中的多人关节姿态跟踪是一个重要而具有挑战性的问题。在本文中,沿着自顶向下的方法的道路,我们提出了一个体面的和有效的姿势跟踪器的基础上的姿势流。首先,我们设计了一个在线优化框架来建立跨帧姿势和表单姿势流的关联(PF-Builder)。其次,设计了一种新的姿态流非最大值抑制算法(PF-NMS)来鲁棒地减少冗余姿态流并重新连接时间不相交的姿态流。大量的实验表明,我们的方法显着优于最好的报告结果在两个标准的姿态跟踪数据集分别为13 mAP 25 MOTA和6 mAP 3 MOTA。此外,在处理各个帧中检测到的姿态的情况下,姿态跟踪器的额外计算非常小,保证在线10 FPS跟踪。我们的源代码是公开的(https://github.com/YuliangXiu/PoseFlow)。

英文摘要:

Multi-person articulated pose tracking in unconstrained videos is an important while challenging problem. In this paper, going along the road of top-down approaches, we propose a decent and efficient pose tracker based on pose flows. First, we design an online optimization framework to build the association of cross-frame poses and form pose flows (PF-Builder). Second, a novel pose flow non-maximum suppression (PF-NMS) is designed to robustly reduce redundant pose flows and re-link temporal disjoint ones. Extensive experiments show that our method significantly outperforms best-reported results on two standard Pose Tracking datasets by 13 mAP 25 MOTA and 6 mAP 3 MOTA respectively. Moreover, in the case of working on detected poses in individual frames, the extra computation of pose tracker is very minor, guaranteeing online 10FPS tracking. Our source codes are made publicly available(https://github.com/YuliangXiu/PoseFlow).

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。

↑