A Personalized Data-Driven Generative Model of Human Motion
Journal:
arXiv
Published Date:
Mar 19, 2025
Abstract
The deployment of autonomous virtual avatars (in extended reality) and robots
in human group activities - such as rehabilitation therapy, sports, and
manufacturing - is expected to increase as these technologies become more
pervasive. Designing cognitive architectures and control strategies to drive
these agents requires realistic models of human motion. However, existing
models only provide simplified descriptions of human motor behavior. In this
work, we propose a fully data-driven approach, based on Long Short-Term Memory
neural networks, to generate original motion that captures the unique
characteristics of specific individuals. We validate the architecture using
real data of scalar oscillatory motion. Extensive analyses show that our model
effectively replicates the velocity distribution and amplitude envelopes of the
individual it was trained on, remaining different from other individuals, and
outperforming state-of-the-art models in terms of similarity to human data.
