Transformers for Enhanced Multi-Object Tracking with Sensor Fusion

Object tracking is a critical component of au-tonomous driving systems, enabling vehicles to understand and predict the movements of surrounding objects. In this paper, we propose an enhanced multi-object tracking framework that integrates transformer-based object detection and sensor fusion. Our approach includes transformer-based 2D detection, fusion and association using transformers, and transformer-based track management. We evaluate our method on the KITTI and nuScenes datasets, demonstrating the performance over baseline methods.