Multi-Camera Multiple 3D Object Tracking on the Move for Autonomous Vehicles
Multi-Camera Multiple 3D Object Tracking on the Move for Autonomous Vehicles
复制标题
DOI：
10.1109/cvprw56347.2022.00289
复制
发表时间：
2022-04
期刊：
影响因子：
--
通讯作者：
Pha Nguyen;Kha Gia Quach;C. Duong;Ngan T. H. Le;Xuan-Bac Nguyen;Khoa Luu
中科院分区：
文献类型：
--
作者：
Pha Nguyen;Kha Gia Quach;C. Duong;Ngan T. H. Le;Xuan-Bac Nguyen;Khoa Luu
自动驾驶汽车的发展提供了一个机会，让一整套摄像头传感器捕捉汽车周围的环境。因此，对象检测和跟踪应对新的挑战非常重要，例如在摄像机视图中实现一致的结果。为了解决这些挑战，这项工作提出了一种新的具有链接预测方法的全局关联图模型，通过交叉注意运动建模和外观重新识别来预测现有轨迹位置和链接检测。该方法旨在解决 3D 对象检测不一致引起的问题。此外，我们的模型在 nuScenes 检测挑战中提高了标准 3D 对象检测器的检测精度。 nuScenes 数据集上的实验结果证明了所提出的方法在现有基于视觉的跟踪数据集上产生 SOTA 性能的好处。
The development of autonomous vehicles provides an opportunity to have a complete set of camera sensors capturing the environment around the car. Thus, it is important for object detection and tracking to address new challenges, such as achieving consistent results across views of cameras. To address these challenges, this work presents a new Global Association Graph Model with Link Prediction approach to predict existing tracklets location and link detections with tracklets via cross-attention motion modeling and appearance re-identification. This approach aims at solving issues caused by inconsistent 3D object detection. Moreover, our model exploits to improve the detection ac-curacy of a standard 3D object detector in the nuScenes detection challenge. The experimental results on the nuScenes dataset demonstrate the benefits of the proposed method to produce SOTA performance on the existing vision-based tracking dataset.