Track to Detect and Segment: An Online Multi-Object Tracker
Track to Detect and Segment: An Online Multi-Object Tracker
复制标题
DOI：
10.1109/cvpr46437.2021.01217
复制
发表时间：
2021-03
期刊：
影响因子：
--
通讯作者：
Jialian Wu;Jiale Cao;Liangchen Song;Yu Wang;Ming Yang;Junsong Yuan
中科院分区：
文献类型：
--
作者：
Jialian Wu;Jiale Cao;Liangchen Song;Yu Wang;Ming Yang;Junsong Yuan
大多数在线多目标跟踪器在没有任何跟踪输入的情况下，在神经网络中独立执行目标检测。在本文中，我们提出了一种新的在线联合检测和跟踪模型，TraDeS (TRAck to DEtect and Segment)，利用跟踪线索来辅助端到端的检测。TraDeS通过成本量推断目标跟踪偏移量，用于传播先前的目标特征，以改进当前目标的检测和分割。在MOT （2D跟踪）、nuScenes （3D跟踪）、MOTS和Youtube-VIS（实例分割跟踪）4个数据集上显示了TraDeS的有效性和优越性。项目页面：https://jialianwu.com/projects/TraDeS.html。
Most online multi-object trackers perform object detection stand-alone in a neural net without any input from tracking. In this paper, we present a new online joint detection and tracking model, TraDeS (TRAck to DEtect and Segment), exploiting tracking clues to assist detection end-to-end. TraDeS infers object tracking offset by a cost volume, which is used to propagate previous object features for improving current object detection and segmentation. Effectiveness and superiority of TraDeS are shown on 4 datasets, including MOT (2D tracking), nuScenes (3D tracking), MOTS and Youtube-VIS (instance segmentation tracking). Project page: https://jialianwu.com/projects/TraDeS.html.