Unknown multiple object tracking using 2D lidar and video camera

Abstract

A multisensor fusion-based unknown object detection and tracking approach is presented. The approach consists of two new algorithms: (i) object segmentation by boundary detection and (ii) data association by integer programming. The proposed approach tightly combines the information from a single-line scanning lidar and a camera. The performance of the proposed algorithm is evaluated by comparing with a state-of-the-art method and is demonstrated with real datasets obtained from a moving platform.