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 Concept Videos

You might also read

Related Articles

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

Sort by
Same author

Prevalence and Landscape of Pathogenic or Likely Pathogenic Germline Variants and Their Association With Somatic Phenotype in Unselected Chinese Patients With Gynecologic Cancers.

JAMA network open·2023
Same author

Determination of Self-(In)compatibility and Inter-(In)compatibility Relationships in Citrus Using Manual Pollination, Microscopy, and S-Genotype Analyses.

Journal of visualized experiments : JoVE·2023
Same author

χ<sup>(2)</sup> nonlinear photonics in integrated microresonators.

Frontiers of optoelectronics·2023
Same author

Risk-reducing salpingo-oophorectomy among Chinese women at increased risk of breast and ovarian cancer.

Journal of ovarian research·2023
Same author

The elevated toxicity of the biodegradation product (guanylurea) from metformin and the antagonistic pattern recognition of combined toxicity: Insight from the pharmaceutical risk assessment and the simulated wastewater treatment.

The Science of the total environment·2023
Same author

Portal vein embolization combined with ex vivo liver resection and autotransplantation: A novel treatment strategy for end-stage and metastatic hepatic alveolar echinococcosis.

Hepatobiliary & pancreatic diseases international : HBPD INT·2023

Related Experiment Video

Updated: Nov 5, 2025

A Method for 3D Reconstruction and Virtual Reality Analysis of Glial and Neuronal Cells
12:49

A Method for 3D Reconstruction and Virtual Reality Analysis of Glial and Neuronal Cells

Published on: September 28, 2019

13.0K

Image-Guided Human Reconstruction via Multi-Scale Graph Transformation Networks.

Kun Li, Hao Wen, Qiao Feng

    IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society
    |May 19, 2021
    PubMed
    Summary

    This study introduces a novel hierarchical graph transformation network for 3D human reconstruction from single images. The method effectively handles pose variations and complex deformations, producing detailed and topologically consistent human models.

    More Related Videos

    Large-scale Reconstructions and Independent, Unbiased Clustering Based on Morphological Metrics to Classify Neurons in Selective Populations
    12:27

    Large-scale Reconstructions and Independent, Unbiased Clustering Based on Morphological Metrics to Classify Neurons in Selective Populations

    Published on: February 15, 2017

    7.1K
    Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
    04:48

    Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

    Published on: July 5, 2024

    594

    Related Experiment Videos

    Last Updated: Nov 5, 2025

    A Method for 3D Reconstruction and Virtual Reality Analysis of Glial and Neuronal Cells
    12:49

    A Method for 3D Reconstruction and Virtual Reality Analysis of Glial and Neuronal Cells

    Published on: September 28, 2019

    13.0K
    Large-scale Reconstructions and Independent, Unbiased Clustering Based on Morphological Metrics to Classify Neurons in Selective Populations
    12:27

    Large-scale Reconstructions and Independent, Unbiased Clustering Based on Morphological Metrics to Classify Neurons in Selective Populations

    Published on: February 15, 2017

    7.1K
    Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
    04:48

    Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

    Published on: July 5, 2024

    594

    Area of Science:

    • Computer Vision
    • 3D Graphics
    • Machine Learning

    Background:

    • 3D human reconstruction from single images is challenging due to pose variations and topological inconsistencies.
    • Existing methods struggle with accurate inference of clothed human models across diverse poses.

    Purpose of the Study:

    • To propose an efficient and effective method for 3D clothed human reconstruction from single images.
    • To address limitations in handling large deformations and maintaining consistent topologies.
    • To introduce a new dataset for training and evaluating 3D human reconstruction models.

    Main Methods:

    • Utilizing a hierarchical graph transformation network for 3D human shape representation.
    • Employing a vertex-based deformation representation to manage large deformations and avoid geometric distortion.
    • Incorporating perceptual image features via a perspective projection layer for enhanced mesh consistency.

    Main Results:

    • The proposed method achieves efficient training and fast convergence with short testing times.
    • Experimental results show superior plausible and complete 3D human reconstruction compared to state-of-the-art approaches.
    • The D2Human dataset provides valuable resources for training and evaluating deep learning frameworks.

    Conclusions:

    • The developed hierarchical graph transformation network offers an effective solution for single-image 3D human reconstruction.
    • The vertex-based deformation representation and perspective projection layer contribute to robust handling of pose and geometry.
    • The D2Human dataset facilitates advancements in dynamic and detailed 3D human modeling.