Multi-camera multi-object tracking on the move via single-stage global association approach

The development of autonomous vehicles generates a tremendous demand for a low-cost solution based on a complete set of camera sensors to perceive the environment around the car. Towards this solution, it is essential for object detection and tracking to address new challenges in multi-camera settings. To address these challenges, this work introduces novel Single-Stage Global Association Tracking approaches to associate one or more detections from multi-cameras with tracked objects. These approaches aim to solve fragment-tracking issues caused by inconsistent 3D object detection. Moreover, our models also improve the detection accuracy of the standard vision-based 3D object detectors in the nuScenes detection challenge. The extensive experimental results on the nuScenes dataset demonstrate the benefits of the proposed method, which outperforms prior vision-based tracking methods in multi-camera settings. •An efficient Fractional Optimal Transport Assignment method to solve one-to-many assignment problem.•A novel Single-Stage Global Assignment method to match one tracked object with multiple corresponding detections in multi-camera overlapping regions.•A novel End-to-end MC-MOT Learning network that performs detection, tracking and association across cameras in a calibrated system.