Lifted Disjoint Paths with Application in Multiple Object Tracking

TL;DR

This paper introduces the lifted disjoint paths problem, an extension with lifted edges for better path connectivity, and applies it to multiple object tracking, achieving state-of-the-art results by preventing ID switches and improving re-identification.

Contribution

It formulates the lifted disjoint paths problem, proves its NP-hardness, and develops LP-relaxation and cutting plane algorithms for practical optimization in multiple object tracking.

Findings

Achieves nearly optimal tracking assignments.

Improves performance on MOT challenge benchmarks.

Effectively prevents ID switches and enhances re-identification.

Abstract

We present an extension to the disjoint paths problem in which additional \emph{lifted} edges are introduced to provide path connectivity priors. We call the resulting optimization problem the lifted disjoint paths problem. We show that this problem is NP-hard by reduction from integer multicommodity flow and 3-SAT. To enable practical global optimization, we propose several classes of linear inequalities that produce a high-quality LP-relaxation. Additionally, we propose efficient cutting plane algorithms for separating the proposed linear inequalities. The lifted disjoint path problem is a natural model for multiple object tracking and allows an elegant mathematical formulation for long range temporal interactions. Lifted edges help to prevent id switches and to re-identify persons. Our lifted disjoint paths tracker achieves nearly optimal assignments with respect to input detections. As a consequence, it leads on all three main benchmarks of the MOT challenge, improving significantly over state-of-the-art.