BiHMP-GAN: Bidirectional 3D Human Motion Prediction GAN

TL;DR

BiHMP-GAN introduces a probabilistic, bidirectional GAN framework for human motion prediction that generates multiple plausible future sequences conditioned on initial poses, improving diversity and accuracy over prior methods.

Contribution

The paper proposes a novel bidirectional GAN architecture with a recursive prediction strategy and a discriminator that regresses an extrinsic factor, enhancing stochastic human motion prediction.

Findings

Outperforms previous methods in qualitative and quantitative evaluations.

Effectively generates diverse human motion sequences conditioned on initial poses.

Demonstrates superior prediction accuracy using critic and action classifier assessments.

Abstract

Human motion prediction model has applications in various fields of computer vision. Without taking into account the inherent stochasticity in the prediction of future pose dynamics, such methods often converges to a deterministic undesired mean of multiple probable outcomes. Devoid of this, we propose a novel probabilistic generative approach called Bidirectional Human motion prediction GAN, or BiHMP-GAN. To be able to generate multiple probable human-pose sequences, conditioned on a given starting sequence, we introduce a random extrinsic factor r, drawn from a predefined prior distribution. Furthermore, to enforce a direct content loss on the predicted motion sequence and also to avoid mode-collapse, a novel bidirectional framework is incorporated by modifying the usual discriminator architecture. The discriminator is trained also to regress this extrinsic factor r, which is used alongside with the intrinsic factor (encoded starting pose sequence) to generate a particular pose sequence. To further regularize the training, we introduce a novel recursive prediction strategy. In spite of being in a probabilistic framework, the enhanced discriminator architecture allows predictions of an intermediate part of pose sequence to be used as a conditioning for prediction of the latter part of the sequence. The bidirectional setup also provides a new direction to evaluate the prediction quality against a given test sequence. For a fair assessment of BiHMP-GAN, we report performance of the generated motion sequence using (i) a critic model trained to discriminate between real and fake motion sequence, and (ii) an action classifier trained on real human motion dynamics. Outcomes of both qualitative and quantitative evaluations, on the probabilistic generations of the model, demonstrate the superiority of BiHMP-GAN over previously available methods.