Real-Time 3D Multi-Object Tracking

Code: RealTimeTracking GitHub Repository

This is a real-time 3D multi-object tracking system based on the AB3DMOT project (GitHub Repository for Tracking). It utilizes lidar point cloud data for 3D object detection, then employs a 3D Kalman filter and the Hungarian algorithm for state estimation and data association, achieving effective object tracking.

Due to limitations in the original code, which restricted it to recorded data, we've adapted the Kalman Filter and data association algorithms to enable real-time applications, such as in self-driving cars.