BASE: Probably a Better Approach to Visual Multi-Object Tracking

Martin Larsen, Martin Larsen, Sigmund Rolfsjord, Sigmund Rolfsjord, Daniel Gusland, Daniel Gusland, Jörgen Ahlberg, Jörgen Ahlberg, Kim Mathiassen, Kim Mathiassen

2024

Abstract

The field of visual object tracking is dominated by methods that combine simple tracking algorithms and ad hoc schemes. Probabilistic tracking algorithms, which are leading in other fields, are surprisingly absent from the leaderboards. We found that accounting for distance in target kinematics, exploiting detector confidence and modelling non-uniform clutter characteristics is critical for a probabilistic tracker to work in visual tracking. Previous probabilistic methods fail to address most or all these aspects, which we believe is why they fall so far behind current state-of-the-art (SOTA) methods (there are no probabilistic trackers in the MOT17 top 100). To rekindle progress among probabilistic approaches, we propose a set of pragmatic models addressing these challenges, and demonstrate how they can be incorporated into a probabilistic framework. We present BASE (Bayesian Approximation Single-hypothesis Estimator), a simple, performant and easily extendible visual tracker, achieving state-of-the-art (SOTA) on MOT17 and MOT20, without using Re-Id. Code available at https://github.com/ffi-no/paper-base-visapp-2024.

Paper Citation

in Harvard Style

Larsen M., Rolfsjord S., Gusland D., Ahlberg J. and Mathiassen K. (2024). BASE: Probably a Better Approach to Visual Multi-Object Tracking. In Proceedings of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 4: VISAPP; ISBN 978-989-758-679-8, SciTePress, pages 110-121. DOI: 10.5220/0012386600003660

in Bibtex Style

@conference{visapp24,
author={Martin Larsen and Sigmund Rolfsjord and Daniel Gusland and Jörgen Ahlberg and Kim Mathiassen},
title={BASE: Probably a Better Approach to Visual Multi-Object Tracking},
booktitle={Proceedings of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 4: VISAPP},
year={2024},
pages={110-121},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012386600003660},
isbn={978-989-758-679-8},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 4: VISAPP
TI - BASE: Probably a Better Approach to Visual Multi-Object Tracking
SN - 978-989-758-679-8
AU - Larsen M.
AU - Rolfsjord S.
AU - Gusland D.
AU - Ahlberg J.
AU - Mathiassen K.
PY - 2024
SP - 110
EP - 121
DO - 10.5220/0012386600003660
PB - SciTePress