CovarianceNet: Conditional Generative Model for Correct Covariance Prediction in Human Motion Prediction

TL;DR

CovarianceNet is a novel conditional generative model that accurately predicts the uncertainty in human motion trajectories by estimating covariance, improving overconfident predictions of existing methods for safer planning and decision-making.

Contribution

The paper introduces CovarianceNet, a new method that predicts covariance in human motion, addressing overconfidence issues in existing models and enhancing uncertainty estimation.

Findings

CovarianceNet accurately predicts uncertainty in human motion trajectories.

It outperforms state-of-the-art methods in uncertainty calibration.

The approach is validated on the ETH dataset.

Abstract

The correct characterization of uncertainty when predicting human motion is equally important as the accuracy of this prediction. We present a new method to correctly predict the uncertainty associated with the predicted distribution of future trajectories. Our approach, CovariaceNet, is based on a Conditional Generative Model with Gaussian latent variables in order to predict the parameters of a bi-variate Gaussian distribution. The combination of CovarianceNet with a motion prediction model results in a hybrid approach that outputs a uni-modal distribution. We will show how some state of the art methods in motion prediction become overconfident when predicting uncertainty, according to our proposed metric and validated in the ETH data-set \cite{pellegrini2009you}. CovarianceNet correctly predicts uncertainty, which makes our method suitable for applications that use predicted distributions, e.g., planning or decision making.