An Online Learning-based Framework for Tracking

TL;DR

This paper introduces an online learning-based framework for object tracking that is more robust to model mismatches than traditional Bayesian methods, demonstrated through improved performance on simulated data.

Contribution

The paper proposes a novel online learning framework for tracking, offering an efficient algorithm that enhances robustness against model mismatches compared to standard Bayesian approaches.

Findings

Our algorithm outperforms Bayesian methods with slight model mismatches.

Experimental results show improved tracking accuracy in simulations.

The framework provides a new perspective on robust object tracking.

Abstract

We study the tracking problem, namely, estimating the hidden state of an object over time, from unreliable and noisy measurements. The standard framework for the tracking problem is the generative framework, which is the basis of solutions such as the Bayesian algorithm and its approximation, the particle filters. However, these solutions can be very sensitive to model mismatches. In this paper, motivated by online learning, we introduce a new framework for tracking. We provide an efficient tracking algorithm for this framework. We provide experimental results comparing our algorithm to the Bayesian algorithm on simulated data. Our experiments show that when there are slight model mismatches, our algorithm outperforms the Bayesian algorithm.