Learning Variations in Human Motion via Mix-and-Match Perturbation

Abstract

Human motion prediction is a stochastic process: Given an observed sequence of poses, multiple future motions are plausible. Existing approaches to modeling this stochasticity typically combine a random noise vector with information about the previous poses. This combination, however, is done in a deterministic manner, which gives the network the flexibility to learn to ignore the random noise. In this paper, we introduce an approach to stochastically combine the root of variations with previous pose information, which forces the model to take the noise into account. We exploit this idea for motion prediction by incorporating it into a recurrent encoder-decoder network with a conditional variational autoencoder block that learns to exploit the perturbations. Our experiments demonstrate that our model yields high-quality pose sequences that are much more diverse than those from state-of-the-art stochastic motion prediction techniques.

DLow: Diversifying Latent Flows for Diverse Human Motion Prediction

“…(1) Average Pairwise Distance (APD): average L2 distance between all pairs of motion samples to measure diversity within samples, which is computed as In these metrics, APD has been used to measure sample diversity [3]. ADE and FDE are common metrics for evaluating sample accuracy in trajectory forecasting literature [2,43,26].…”

“…More related to our work, stochastic human motion prediction methods start to gain popularity with the development of deep generative models. These methods [68,46,6,61,42,71,3] often build upon popular generative models such as conditional generative adversarial networks (CGANs; [21]) or conditional variational autoencoders (CVAEs; [38]). The aforementioned methods differ in the design of their generative models, but at test time they follow the same sampling strategy -randomly and independently sampling trajectories from the pretrained generative model without considering the correlation between samples.…”

“…Deep generative models, e.g., variational autoencoders (VAEs) [38], are effective tools to model multi-modal data distributions. Most existing work [68,46,6,61,42,71,3] using deep generative models for human motion prediction is focused on the design of the generative model to allow it to effectively learn the data distribution. After the generative model is learned, little attention has been paid to the sampling method used to produce motion samples (predicted future motions) from the pretrained generative model (weights kept fixed).…”