Overlapping Mixtures of Gaussian Processes for the Data Association Problem

TL;DR

This paper introduces a novel mixture of Gaussian Processes that clusters data based on trajectories without gating functions, improving data association in tracking and regression tasks.

Contribution

It presents a new GP mixture model with no gating function, using a variational Bayesian algorithm for efficient label recovery and hyperparameter learning.

Findings

Effective in multi-object tracking disambiguation

Applicable to traditional regression problems

Demonstrates improved data association accuracy

Abstract

In this work we introduce a mixture of GPs to address the data association problem, i.e. to label a group of observations according to the sources that generated them. Unlike several previously proposed GP mixtures, the novel mixture has the distinct characteristic of using no gating function to determine the association of samples and mixture components. Instead, all the GPs in the mixture are global and samples are clustered following "trajectories" across input space. We use a non-standard variational Bayesian algorithm to efficiently recover sample labels and learn the hyperparameters. We show how multi-object tracking problems can be disambiguated and also explore the characteristics of the model in traditional regression settings.