Probabilistic 3D Multi-Object Tracking for Autonomous Driving

TL;DR

This paper introduces a probabilistic 3D multi-object tracking method for autonomous driving that uses Kalman Filters and outperforms existing baselines, winning the NuScenes Tracking Challenge.

Contribution

The paper presents a novel online 3D multi-object tracking approach utilizing Kalman Filters with statistical initialization, achieving state-of-the-art results in autonomous driving scenarios.

Findings

Outperforms AB3DMOT baseline significantly in AMOTA metric.

Wins first place in NuScenes Tracking Challenge.

Demonstrates effective use of Kalman Filter for 3D object tracking.

Abstract

3D multi-object tracking is a key module in autonomous driving applications that provides a reliable dynamic representation of the world to the planning module. In this paper, we present our on-line tracking method, which made the first place in the NuScenes Tracking Challenge, held at the AI Driving Olympics Workshop at NeurIPS 2019. Our method estimates the object states by adopting a Kalman Filter. We initialize the state covariance as well as the process and observation noise covariance with statistics from the training set. We also use the stochastic information from the Kalman Filter in the data association step by measuring the Mahalanobis distance between the predicted object states and current object detections. Our experimental results on the NuScenes validation and test set show that our method outperforms the AB3DMOT baseline method by a large margin in the Average Multi-Object Tracking Accuracy (AMOTA) metric.