Parametric annealing: a stochastic search method for human pose tracking

TL;DR

This paper introduces a parametric annealing method that enhances global optimization for human pose tracking, reducing computational costs and improving accuracy over existing techniques like APF.

Contribution

It proposes a novel annealing approach that reuses samples and employs adaptive parametric densities, outperforming APF in efficiency and accuracy.

Findings

Reduces tracking error compared to APF

Uses less than 50% of the computational resources of APF

Shows qualitative improvements in pose tracking results

Abstract

Model based methods to marker-free motion capture have a very high computational overhead that make them unattractive. In this paper we describe a method that improves on existing global optimization techniques to tracking articulated objects. Our method improves on the state-of-the-art Annealed Particle Filter (APF) by reusing samples across annealing layers and by using an adaptive parametric density for diffusion. We compare the proposed method with APF on a scalable problem and study how the two methods scale with the dimensionality, multi-modality and the range of search. Then we perform sensitivity analysis on the parameters of our algorithm and show that it tolerates a wide range of parameter settings. We also show results on tracking human pose from the widely-used Human Eva I dataset. Our results show that the proposed method reduces the tracking error despite using less than 50% of the computational resources as APF. The tracked output also shows a significant qualitative improvement over APF as demonstrated through image and video results.