Here is a novel approach to recreating athletic movements in 3D.
Check out a paper on a framework for simulating different athletic skills such as the high jump. "The strategies are realized as control policies for physics-based characters," reads the abstract. "Given a task objective and an initial character configuration, the combination of physics simulation and deep reinforcement learning (DRL) provides a suitable starting point for automatic control policy training."
The team explored using a Pose Variational Autoencoder (P-VAE) to constrain the actions to a subspace of natural poses. "In contrast to motion imitation methods, a rich variety of novel strategies can naturally emerge by exploring initial character states through a sample-efficient Bayesian diversity search (BDS) algorithm," noted the team.
The researchers state that the new method offers more realistic results and more flexibility. You can find the full paper here.
Don't forget to join our new Telegram channel, our Discord, follow us on Instagram and Twitter, where we are sharing breakdowns, the latest news, awesome artworks, and more.