Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Gaussian process dynamical models for human motion.

Jack M Wang1, David J Fleet, Aaron Hertzmann

  • 1Department of Computer Science, University of Toronto, 40 St. George Street, Toronto, Ontario M5S 2E4 Canada. jmwang@dgp.toronto.edu

IEEE Transactions on Pattern Analysis and Machine Intelligence
|December 18, 2007
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Lossy processing principles in 2D and 3D vision.

The Behavioral and brain sciences·2025
Same author

Viewers perceive shape in pictures according to per-fixation perspective.

Scientific reports·2025
Same author

Benchmark suites instead of leaderboards for evaluating AI fairness.

Patterns (New York, N.Y.)·2024
Same author

Personalized Video-Based Hand Taxonomy Using Egocentric Video in the Wild.

IEEE journal of biomedical and health informatics·2024
Same author

Toward a theory of perspective perception in pictures.

Journal of vision·2024
Same author

Deep generative priors for biomolecular 3D heterogeneous reconstruction from cryo-EM projections.

Journal of structural biology·2024
Same journal

Relation DETR+: Exploring Explicit Position Relation Prior for Dense Prediction.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

RBF++: Quantifying and Optimizing Reasoning Boundaries across Measurable and Unmeasurable Capabilities for Chain-of-Thought Reasoning.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

CAFE: Cross-View Adaptive Fusion and Cluster Center Enhancement for Robust Multi-View Clustering.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

DIVER: Reinforced Diffusion Breaks Imitation Bottlenecks in End-to-End Autonomous Driving.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Ethics-Aware Safe Reinforcement Learning for Rare-Event Risk Control in Interactive Urban Driving.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Learning Shape Anchors for Holistic Indoor Scene Understanding.

IEEE transactions on pattern analysis and machine intelligence·2026
See all related articles

Gaussian Process Dynamical Models (GPDM) offer a novel approach to nonlinear time series analysis. This method effectively models complex human motion dynamics from high-dimensional motion capture data, even with limited datasets.

Area of Science:

  • Machine Learning
  • Dynamical Systems
  • Time Series Analysis

Background:

  • Nonlinear time series analysis is crucial for understanding complex systems.
  • High-dimensional motion capture data presents challenges for traditional modeling techniques.
  • Existing models often struggle to capture the inherent uncertainty in dynamical systems.

Purpose of the Study:

  • To introduce Gaussian Process Dynamical Models (GPDM) for nonlinear time series analysis.
  • To apply GPDM to learning models of human pose and motion from motion capture data.
  • To develop a non-parametric dynamical system model that accounts for uncertainty.

Main Methods:

  • Developed a latent variable model comprising a low-dimensional latent space with dynamics and a mapping to observation space.

Related Experiment Videos

  • Utilized Gaussian process priors for dynamics and observation mappings.
  • Marginalized model parameters in closed-form to create a non-parametric model.

    • Main Results:

    • Demonstrated the effectiveness of GPDM on high-dimensional human motion capture data (50-dimensional poses).
    • Compared four distinct learning algorithms for GPDM.
    • Showcased successful learning of nonlinear dynamics despite small dataset sizes.

    Conclusions:

    • GPDM provides an effective non-parametric approach for modeling nonlinear dynamical systems.
    • The model successfully captures complex human motion dynamics from motion capture data.
    • GPDM offers a robust method for time series analysis with inherent uncertainty.