Learning to predict diverse human motions from a single image via mixture density networks

Human motion prediction, which plays a key role in computer vision, generally requires a past motion sequence as input. However, in real applications, a complete and correct past motion sequence can be too expensive to achieve. In this paper, we propose a novel approach to predicting future human motions from a much weaker condition, i.e., a single image, with mixture density networks (MDN) modeling. Contrary to most existing deep human motion prediction approaches, the multimodal nature of MDN enables the generation of diverse future motion hypotheses, which well compensates for the strong stochastic ambiguity aggregated by the single input and human motion uncertainty. In designing the loss function, we further introduce the energy-based formulation to flexibly impose prior losses over the learnable parameters of MDN to maintain motion coherence as well as improve the prediction accuracy by customizing the energy functions. Our trained model directly takes an image as input and generates multiple plausible motions that satisfy the given condition. Extensive experiments on two standard benchmark datasets demonstrate the effectiveness of our method in terms of prediction diversity and accuracy. (C) 2022 Elsevier B.V. All rights reserved.

Automated summary

In this paper, a novel approach based on mixture density networks (MDN) is proposed to predict future human motions from a single image. Unlike existing methods, the multimodal nature of MDN allows the generation of diverse motion hypotheses, compensating for the stochastic ambiguity caused by the single input and human motion uncertainty. The energy-based formulation is introduced to customize the energy functions and improve the prediction accuracy while maintaining motion coherence. Experimental results on benchmark datasets demonstrate the effectiveness of the proposed method in terms of prediction diversity and accuracy.