Tracking Task

In 3D object tracking for point clouds, we provide the IDs and bounding box labels for each object in a sequence of scenes. The object ID is unique and consistent in time. The tracking method outputs a label and ID for each object in a sequence of scenes.

Metric

We use traditional Multi-Object Tracking Accuracy (MOTA) to measure the tracking result: MOTA = 1-(FP+IDS+FN)/GT where FP and FN are false positive and false negative, IDS denotes the false ID matching for tracking in different time-steps, and GT is the number of ground truth tracked instances.

ML(mostly loss) is the proportion of successful track matching of ground truth in less than 20% of the time in all tracking targets.

MT(mostly track) is the proportion of successful track matching of ground truth in more than 80% of the time in all tracking targets.

TASKS Icon