A survey of human gait-based artificial intelligence applications

EJ Harris, IH Khoo, E Demircan - Frontiers in Robotics and AI, 2022 - frontiersin.org
We performed an electronic database search of published works from 2012 to mid-2021 that
focus on human gait studies and apply machine learning techniques. We identified six key …

Recent advances in imitation learning from observation

F Torabi, G Warnell, P Stone - arXiv preprint arXiv:1905.13566, 2019 - arxiv.org
Imitation learning is the process by which one agent tries to learn how to perform a certain
task using information generated by another, often more-expert agent performing that same …

Physdiff: Physics-guided human motion diffusion model

Y Yuan, J Song, U Iqbal, A Vahdat… - Proceedings of the …, 2023 - openaccess.thecvf.com
Denoising diffusion models hold great promise for generating diverse and realistic human
motions. However, existing motion diffusion models largely disregard the laws of physics in …

Humans in 4D: Reconstructing and tracking humans with transformers

S Goel, G Pavlakos, J Rajasegaran… - Proceedings of the …, 2023 - openaccess.thecvf.com
We present an approach to reconstruct humans and track them over time. At the core of our
approach, we propose a fully" transformerized" version of a network for human mesh …

Video pretraining (vpt): Learning to act by watching unlabeled online videos

B Baker, I Akkaya, P Zhokov… - Advances in …, 2022 - proceedings.neurips.cc
Pretraining on noisy, internet-scale datasets has been heavily studied as a technique for
training models with broad, general capabilities for text, images, and other modalities …

Physical inertial poser (pip): Physics-aware real-time human motion tracking from sparse inertial sensors

X Yi, Y Zhou, M Habermann… - Proceedings of the …, 2022 - openaccess.thecvf.com
Motion capture from sparse inertial sensors has shown great potential compared to image-
based approaches since occlusions do not lead to a reduced tracking quality and the …

Humor: 3d human motion model for robust pose estimation

D Rempe, T Birdal, A Hertzmann… - Proceedings of the …, 2021 - openaccess.thecvf.com
We introduce HuMoR: a 3D Human Motion Model for Robust Estimation of temporal pose
and shape. Though substantial progress has been made in estimating 3D human motion …

Amp: Adversarial motion priors for stylized physics-based character control

XB Peng, Z Ma, P Abbeel, S Levine… - ACM Transactions on …, 2021 - dl.acm.org
Synthesizing graceful and life-like behaviors for physically simulated characters has been a
fundamental challenge in computer animation. Data-driven methods that leverage motion …

Reinforcement learning with action-free pre-training from videos

Y Seo, K Lee, SL James… - … Conference on Machine …, 2022 - proceedings.mlr.press
Recent unsupervised pre-training methods have shown to be effective on language and
vision domains by learning useful representations for multiple downstream tasks. In this …

Learning agile robotic locomotion skills by imitating animals

XB Peng, E Coumans, T Zhang, TW Lee, J Tan… - arXiv preprint arXiv …, 2020 - arxiv.org
Reproducing the diverse and agile locomotion skills of animals has been a longstanding
challenge in robotics. While manually-designed controllers have been able to emulate many …