End-to-end Learning of Multi-sensor 3D Tracking by Detection
This addresses the problem of accurate 3D trajectory estimation for autonomous driving, but it is incremental as it builds on existing tracking-by-detection methods.
The paper tackles 3D object tracking by detection using multi-sensor data from cameras and LIDAR, achieving very competitive results on the KITTI dataset.
In this paper we propose a novel approach to tracking by detection that can exploit both cameras as well as LIDAR data to produce very accurate 3D trajectories. Towards this goal, we formulate the problem as a linear program that can be solved exactly, and learn convolutional networks for detection as well as matching in an end-to-end manner. We evaluate our model in the challenging KITTI dataset and show very competitive results.