Multi Target Tracking by Learning from Generalized Graph Differences

TL;DR

This paper introduces a novel learning approach for multi-object tracking that leverages generalized graph differences to efficiently train network weights, improving tracking performance without complex inner-loop optimizations.

Contribution

It proposes a new method that separates embedding and optimization, using graph differences for training data generation, enhancing multi-target tracking accuracy.

Findings

Achieved competitive results on DukeMTMCT dataset.

Efficient training without additional inner-loop optimizations.

Effective use of generalized graph differences for data representation.

Abstract

Formulating the multi object tracking problem as a network flow optimization problem is a popular choice. In this paper an efficient way of learning the weights of such a network is presented. It separates the problem into one embedding of feasible solutions into a one dimensional feature space and one optimization problem. The embedding can be learned using standard SGD type optimization without relying on an additional optimizations within each step. Training data is produced by performing small perturbations of ground truth tracks and representing them using generalized graph differences, which is an efficient way introduced to represent the difference between two graphs. The proposed method is evaluated on DukeMTMCT with competitive results.