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

ImageNet3D: Towards General-Purpose Object-Level 3D Understanding.

Advances in neural information processing systems·2026
Same author

Proximity-Driven Protein Ligation Beyond the Concentration Limit.

Journal of the American Chemical Society·2026
Same author

Proactive Health: A Culture-Centered Study on the Differential Health Practices of Older Adults in Elderly Care Institutions in China.

Health communication·2026
Same author

Direct Photoredox Synthesis of <i>N</i>-Linked Glycoproteins.

Journal of the American Chemical Society·2026
Same author

Elemental phosphorus-stabilized Ni<sub>2</sub>P for efficient electrooxidation of 5-hydroxymethylfurfural to 2,5-furandicarboxylic acid.

Chemical communications (Cambridge, England)·2026
Same author

Evaluation of a novel C2 pedicle screw insertion technique: a retrospective comparative clinical study and finite element analysis.

Scientific reports·2026

Related Experiment Video

Updated: Apr 24, 2026

Photorealistic Learned Landscapes for Augmented Reality
06:54

Photorealistic Learned Landscapes for Augmented Reality

Published on: June 27, 2025

848

Robust 3D reconstruction with an RGB-D camera.

Kangkan Wang, Guofeng Zhang, Hujun Bao

    IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society
    |September 10, 2014
    PubMed
    Summary

    This study introduces a new 3D reconstruction method using affordable RGB-D cameras. It effectively handles challenging scans with texture repetition and missing depth data, improving 3D model accuracy.

    Area of Science:

    • Computer Vision
    • 3D Reconstruction
    • Robotics

    Background:

    • Accurate 3D reconstruction is crucial for various applications, but challenges like repetitive textures and missing depth data hinder current methods.
    • Existing scanning systems struggle with robustness and efficiency, especially in complex environments.

    Purpose of the Study:

    • To develop a novel and robust 3D reconstruction approach using low-cost RGB-D cameras.
    • To overcome limitations of existing methods in handling challenging scan data, including repetitive textures and significant depth missing.

    Main Methods:

    • Utilizing a combination of visual and geometry features for robust registration.
    • Employing Structure from Motion (SFM) techniques to enhance feature matching and camera pose estimation.
    • Introducing a novel prior-based multi-candidates RANSAC for efficient parameter estimation and accelerated camera pose estimation.

    More Related Videos

    Robotized Testing of Camera Positions to Determine Ideal Configuration for Stereo 3D Visualization of Open-Heart Surgery
    05:12

    Robotized Testing of Camera Positions to Determine Ideal Configuration for Stereo 3D Visualization of Open-Heart Surgery

    Published on: August 12, 2021

    1.8K
    Reefshape: A System for the Efficient Collection and Automated Processing of Time-Series Underwater Photogrammetry Data for Benthic Habitat Monitoring
    13:35

    Reefshape: A System for the Efficient Collection and Automated Processing of Time-Series Underwater Photogrammetry Data for Benthic Habitat Monitoring

    Published on: June 13, 2025

    1.9K

    Related Experiment Videos

    Last Updated: Apr 24, 2026

    Photorealistic Learned Landscapes for Augmented Reality
    06:54

    Photorealistic Learned Landscapes for Augmented Reality

    Published on: June 27, 2025

    848
    Robotized Testing of Camera Positions to Determine Ideal Configuration for Stereo 3D Visualization of Open-Heart Surgery
    05:12

    Robotized Testing of Camera Positions to Determine Ideal Configuration for Stereo 3D Visualization of Open-Heart Surgery

    Published on: August 12, 2021

    1.8K
    Reefshape: A System for the Efficient Collection and Automated Processing of Time-Series Underwater Photogrammetry Data for Benthic Habitat Monitoring
    13:35

    Reefshape: A System for the Efficient Collection and Automated Processing of Time-Series Underwater Photogrammetry Data for Benthic Habitat Monitoring

    Published on: June 13, 2025

    1.9K
  • Integrating multiview stereo and mesh deformation for completing missing geometry.

    • Main Results:

    • Successful registration of frames even with substantial depth missing.
    • Robust loop closure detection and handling to mitigate drift.
    • Demonstrated effectiveness in challenging 3D reconstruction scenarios with repetitive textures.
    • Improved accuracy and robustness in camera pose estimation.

    Conclusions:

    • The proposed 3D reconstruction approach offers a robust and efficient solution for low-cost RGB-D scanning systems.
    • The method effectively addresses common challenges in 3D reconstruction, enabling high-quality results in difficult conditions.
    • This work advances the field of 3D reconstruction by providing a practical and effective system for real-world applications.