Generating Smooth Pose Sequences for Diverse Human Motion Prediction

TL;DR

This paper presents a unified deep generative network that produces diverse and controllable human motion sequences by leveraging a smooth pose prior, outperforming existing methods on benchmark datasets.

Contribution

A novel deep generative model that unifies diverse and controllable human motion prediction using a pose prior and sequential part prediction.

Findings

Outperforms state-of-the-art in diversity and accuracy

Effective use of a normalizing flow based pose prior

Demonstrates on Human3.6M and HumanEva-I datasets

Abstract

Recent progress in stochastic motion prediction, i.e., predicting multiple possible future human motions given a single past pose sequence, has led to producing truly diverse future motions and even providing control over the motion of some body parts. However, to achieve this, the state-of-the-art method requires learning several mappings for diversity and a dedicated model for controllable motion prediction. In this paper, we introduce a unified deep generative network for both diverse and controllable motion prediction. To this end, we leverage the intuition that realistic human motions consist of smooth sequences of valid poses, and that, given limited data, learning a pose prior is much more tractable than a motion one. We therefore design a generator that predicts the motion of different body parts sequentially, and introduce a normalizing flow based pose prior, together with a joint angle loss, to achieve motion realism.Our experiments on two standard benchmark datasets, Human3.6M and HumanEva-I, demonstrate that our approach outperforms the state-of-the-art baselines in terms of both sample diversity and accuracy. The code is available at https://github.com/wei-mao-2019/gsps