Unsupervised and Generic Short-Term Anticipation of Human Body Motions

TL;DR

This paper introduces a novel, interpretable, and generic method using Dynamic Mode Decomposition with delays for short-term human motion anticipation, achieving comparable or better accuracy than neural networks with less training.

Contribution

The paper presents a new approach employing Dynamic Mode Decomposition with delays, enhancing interpretability and reducing training time for short-term human motion prediction.

Findings

Anticipation errors are comparable or better for <0.4 sec prediction.

Method is interpretable through linear combinations of factors.

Requires less training time than neural network methods.

Abstract

Various neural network based methods are capable of anticipating human body motions from data for a short period of time. What these methods lack are the interpretability and explainability of the network and its results. We propose to use Dynamic Mode Decomposition with delays to represent and anticipate human body motions. Exploring the influence of the number of delays on the reconstruction and prediction of various motion classes, we show that the anticipation errors in our results are comparable or even better for very short anticipation times ($<0.4$ sec) to a recurrent neural network based method. We perceive our method as a first step towards the interpretability of the results by representing human body motions as linear combinations of ``factors''. In addition, compared to the neural network based methods large training times are not needed. Actually, our methods do not even regress to any other motions than the one to be anticipated and hence is of a generic nature.