AnyPose: Anytime 3D Human Pose Forecasting via Neural Ordinary Differential Equations

TL;DR

AnyPose is a lightweight continuous-time neural network using neural ODEs for accurate, real-time 3D human pose forecasting at any time point, outperforming existing methods in speed and accuracy.

Contribution

The paper introduces AnyPose, a novel neural ODE-based architecture for continuous-time human pose prediction, enabling anytime forecasting with improved efficiency.

Findings

High accuracy in pose prediction on multiple datasets.

Significantly reduced computational time compared to traditional methods.

Effective modeling of human behavior dynamics with neural ODEs.

Abstract

Anytime 3D human pose forecasting is crucial to synchronous real-world human-machine interaction, where the term ``anytime" corresponds to predicting human pose at any real-valued time step. However, to the best of our knowledge, all the existing methods in human pose forecasting perform predictions at preset, discrete time intervals. Therefore, we introduce AnyPose, a lightweight continuous-time neural architecture that models human behavior dynamics with neural ordinary differential equations. We validate our framework on the Human3.6M, AMASS, and 3DPW dataset and conduct a series of comprehensive analyses towards comparison with existing methods and the intersection of human pose and neural ordinary differential equations. Our results demonstrate that AnyPose exhibits high-performance accuracy in predicting future poses and takes significantly lower computational time than traditional methods in solving anytime prediction tasks.