The Smooth Trajectory Estimator for LMB Filters

TL;DR

This paper introduces a smooth-trajectory estimator for the LMB filter that leverages the structure of the GLMB filter, improving tracking accuracy with minimal additional computational cost.

Contribution

It presents a novel smooth-trajectory estimator for the LMB filter based on the best association map, enhancing trajectory estimation in multi-object tracking.

Findings

Significant improvements in tracking accuracy demonstrated in experiments.

Negligible increase in computational time compared to standard LMB filter.

Method is effective in challenging tracking scenarios.

Abstract

This paper proposes a smooth-trajectory estimator for the labelled multi-Bernoulli (LMB) filter by exploiting the special structure of the generalised labelled multi-Bernoulli (GLMB) filter. We devise a simple and intuitive approach to store the best association map when approximating the GLMB random finite set (RFS) to the LMB RFS. In particular, we construct a smooth-trajectory estimator (i.e., an estimator over the entire trajectories of labelled estimates) for the LMB filter based on the history of the best association map and all of the measurements up to the current time. Experimental results under two challenging scenarios demonstrate significant tracking accuracy improvements with negligible additional computational time compared to the conventional LMB filter. The source code is publicly available at https://tinyurl.com/ste-lmb, aimed at promoting advancements in MOT algorithms.