Loading [a11y]/accessibility-menu.js
DFR-FastMOT: Detection Failure Resistant Tracker for Fast Multi-Object Tracking Based on Sensor Fusion | IEEE Conference Publication | IEEE Xplore

DFR-FastMOT: Detection Failure Resistant Tracker for Fast Multi-Object Tracking Based on Sensor Fusion


Abstract:

Persistent multi-object tracking (MOT) allows autonomous vehicles to navigate safely in highly dynamic environments. One of the well-known challenges in MOT is object occ...Show More

Abstract:

Persistent multi-object tracking (MOT) allows autonomous vehicles to navigate safely in highly dynamic environments. One of the well-known challenges in MOT is object occlusion when an object becomes unobservant for subsequent frames. The current MOT methods store objects information, such as trajectories, in internal memory to recover the objects after occlusions. However, they retain short-term memory to save computational time and avoid slowing down the MOT method. As a result, they lose track of objects in some occlusion scenarios, particularly long ones. In this paper, we propose DFR-FastMOT, a light MOT method that uses data from a camera and LiDAR sensors and relies on an algebraic formulation for object association and fusion. The formulation boosts the computational time and permits long-term memory that tackles more occlusion scenarios. Our method shows outstanding tracking performance over recent learning and non-learning benchmarks with about 3% and 4% margin in MOTA, respectively. Also, we conduct extensive experiments that simulate occlusion phenomena by employing detectors with various distortion levels. The proposed solution enables superior performance under various distortion levels in detection over current state-of-art methods. Our framework processes about 7,763 frames in 1.48 seconds, which is seven times faster than recent benchmarks. The framework will be available at https://github.com/MohamedNagyMostafa/DFR-FastMOT.
Date of Conference: 29 May 2023 - 02 June 2023
Date Added to IEEE Xplore: 04 July 2023
ISBN Information:
Conference Location: London, United Kingdom

Funding Agency:


I. Introduction

Multi-object tracking (MOT) provides information about the surrounding objects, allowing autonomous vehicles (AVs) to avoid collisions with other cars by making proper navigation decisions. AVs rely on sensors such as cameras and LiDAR to obtain sufficient information for object tracking. We categorize the recent research work into two groups: the first group uses mono-sensor, and the other group employs multi-sensors to collect information for MOT.

Contact IEEE to Subscribe

References

References is not available for this document.