Learnable Online Graph Representations for 3D Multi-Object Tracking
This addresses the problem of suboptimal performance in 3D MOT for applications like autonomous driving by replacing handcrafted features with a trainable method, though it is incremental as it builds on existing graph-based and learning techniques.
The paper tackles 3D multi-object tracking by proposing a learning-based approach using a graph structure and neural message passing for data association, achieving state-of-the-art performance with 65.6% AMOTA and 58% fewer ID-switches on the nuScenes dataset.
Tracking of objects in 3D is a fundamental task in computer vision that finds use in a wide range of applications such as autonomous driving, robotics or augmented reality. Most recent approaches for 3D multi object tracking (MOT) from LIDAR use object dynamics together with a set of handcrafted features to match detections of objects. However, manually designing such features and heuristics is cumbersome and often leads to suboptimal performance. In this work, we instead strive towards a unified and learning based approach to the 3D MOT problem. We design a graph structure to jointly process detection and track states in an online manner. To this end, we employ a Neural Message Passing network for data association that is fully trainable. Our approach provides a natural way for track initialization and handling of false positive detections, while significantly improving track stability. We show the merit of the proposed approach on the publicly available nuScenes dataset by achieving state-of-the-art performance of 65.6% AMOTA and 58% fewer ID-switches.