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 Video

Updated: Jun 15, 2025

3D Planning and Printing of Patient Specific Implants for Reconstruction of Bony Defects
08:15

3D Planning and Printing of Patient Specific Implants for Reconstruction of Bony Defects

Published on: August 4, 2020

6.4K

Playing for 3D Human Recovery.

Zhongang Cai, Mingyuan Zhang, Jiawei Ren

    IEEE Transactions on Pattern Analysis and Machine Intelligence
    |August 27, 2024
    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

    XOV-Action: Towards Generalizable Open-Vocabulary Action Recognition.

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

    Higher order Magnus expansion for driven two-level quantum dynamics.

    The Journal of chemical physics·2026
    Same author

    CuDi: Curve Distillation for Efficient and Controllable Exposure Adjustment.

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

    The correlation between physical exercise and cognitive function in older adults with cognitive impairment mediated by inhibitory control and possible cognitive processing mechanisms.

    Frontiers in public health·2026
    Same author

    Line-scan phase measuring deflectometry.

    Optics letters·2026
    Same author

    The effects of exercise on generalized anxiety symptoms in adults: a systematic review and meta-analysis.

    BMC sports science, medicine & rehabilitation·2026
    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

    Generating large-scale 3D human recovery datasets using video games like GTA-V proves effective. This synthetic data, with automatic annotations, significantly boosts performance for pose and shape estimation, complementing real-world data limitations.

    Area of Science:

    • Computer Vision
    • Machine Learning
    • Human Pose and Shape Estimation

    Background:

    • 3D human recovery (pose and shape estimation) has advanced significantly.
    • Existing datasets are limited in scale and diversity due to high motion capture costs.

    Purpose of the Study:

    • Introduce GTA-Human, a large-scale 3D human dataset generated using the GTA-V game engine.
    • Investigate the effectiveness and insights gained from using game-playing data for 3D human recovery.

    Main Methods:

    • Generated massive human sequences with automatically annotated 3D ground truths using GTA-V.
    • Trained baseline and advanced models on the GTA-Human dataset.
    • Conducted systematic studies on data scale, domain gap, and model sensitivity.

    More Related Videos

    Author Spotlight: Advancing Human Cardiac Anatomy Through Multi-Scale Analysis of Hearts
    04:22

    Author Spotlight: Advancing Human Cardiac Anatomy Through Multi-Scale Analysis of Hearts

    Published on: June 28, 2024

    410
    3D Printing Model of a Patient's Specific Lumbar Vertebra
    07:30

    3D Printing Model of a Patient's Specific Lumbar Vertebra

    Published on: April 14, 2023

    1.8K

    Related Experiment Videos

    Last Updated: Jun 15, 2025

    3D Planning and Printing of Patient Specific Implants for Reconstruction of Bony Defects
    08:15

    3D Planning and Printing of Patient Specific Implants for Reconstruction of Bony Defects

    Published on: August 4, 2020

    6.4K
    Author Spotlight: Advancing Human Cardiac Anatomy Through Multi-Scale Analysis of Hearts
    04:22

    Author Spotlight: Advancing Human Cardiac Anatomy Through Multi-Scale Analysis of Hearts

    Published on: June 28, 2024

    410
    3D Printing Model of a Patient's Specific Lumbar Vertebra
    07:30

    3D Printing Model of a Patient's Specific Lumbar Vertebra

    Published on: April 14, 2023

    1.8K

    Main Results:

    • A simple baseline trained on GTA-Human outperformed sophisticated methods.
    • Synthetic data effectively complements real-world indoor data, addressing domain gaps.
    • Dataset scale and data density significantly impact model performance.
    • Rich SMPL parameter labels in GTA-Human provide strong supervision.
    • Benefits extend to larger models like CNNs and Transformers.

    Conclusions:

    • Game-generated data is a viable and effective resource for large-scale 3D human recovery.
    • Understanding domain gaps and data mixture strategies is crucial for leveraging synthetic data.
    • The GTA-Human dataset offers a scalable solution for advancing 3D human recovery research.