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 11, 2026

Photorealistic Learned Landscapes for Augmented Reality
06:54

Photorealistic Learned Landscapes for Augmented Reality

Published on: June 27, 2025

A Progressive Gaussian Splatting Framework for Monocular-Only High-Fidelity Surgical Scene Reconstruction.

Yuchao Zheng, Jianing Zhang, Yutong Wu

    IEEE Transactions on Bio-Medical Engineering
    |June 9, 2026
    PubMed
    Summary

    SurGSplat++ offers a pose-free 3D reconstruction method for minimally invasive surgery (MIS), improving geometric stability and enabling better surgical planning and analysis without extra hardware.

    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

    Clinically constrained optical design of a high-numerical aperture miniature immersion objective for probe-based confocal laser endomicroscopy.

    Bioscience trends·2026
    Same author

    Robotic-assisted bronchoscopy in dye localization in thoracoscopic pulmonary nodule resection: An initial experience.

    JTCVS techniques·2026
    Same author

    Noise2Average: An iterative residual learning strategy for image denoising without clean data.

    Imaging neuroscience (Cambridge, Mass.)·2026
    Same author

    RT-SAM: Visual-Prompt Fusion and Uncertainty Enhancement for Nasopharyngeal Carcinoma Radiotherapy Target Delineation.

    IEEE journal of biomedical and health informatics·2026
    Same author

    BCIRT: Backscattering-corrected implicit representation tomography.

    Medical image analysis·2026
    Same author

    Quasi-static Elastography-driven Automated Robotic Ultrasound Screening and Localization.

    IEEE transactions on bio-medical engineering·2025

    Area of Science:

    • Computer Vision
    • Medical Imaging
    • Surgical Robotics

    Background:

    • Accurate 3D scene reconstruction is crucial for intelligent surgery and minimally invasive surgery (MIS).
    • Existing clinical systems are often bulky and workflow-incompatible.
    • Vision-based Structure-from-Motion methods face challenges with sparse textures and specularities, causing unstable pose estimation and high computational costs.

    Purpose of the Study:

    • To develop a novel framework for monocular surgical scene reconstruction that overcomes the limitations of existing methods.
    • To provide a pose-free, progressive Gaussian splatting approach for enhanced 3D scene reconstruction in MIS.
    • To enable accurate and consistent 3D reconstructions without auxiliary hardware or pre-computed camera poses.

    Main Methods:

    More Related Videos

    Medical-grade Sterilizable Target for Fluid-immersed Fetoscope Optical Distortion Calibration
    07:03

    Medical-grade Sterilizable Target for Fluid-immersed Fetoscope Optical Distortion Calibration

    Published on: February 23, 2017

    Related Experiment Videos

    Last Updated: Jun 11, 2026

    Photorealistic Learned Landscapes for Augmented Reality
    06:54

    Photorealistic Learned Landscapes for Augmented Reality

    Published on: June 27, 2025

    Medical-grade Sterilizable Target for Fluid-immersed Fetoscope Optical Distortion Calibration
    07:03

    Medical-grade Sterilizable Target for Fluid-immersed Fetoscope Optical Distortion Calibration

    Published on: February 23, 2017

  • Introduced SurGSplat++, a progressive, pose-free Gaussian splatting framework.
  • Utilized monocular video input for surgical scene reconstruction.
  • Eliminated the need for auxiliary hardware and pre-computed camera poses.

    • Main Results:

    • SurGSplat++ demonstrated improved geometric stability and reduced pose drift compared to existing approaches.
    • Achieved superior novel-view synthesis, enhancing the quality of 3D reconstructions.
    • Validated the practical utility of the method for clinical applications.

    Conclusions:

    • SurGSplat++ offers a practical and efficient solution for monocular surgical scene reconstruction.
    • The framework provides accurate and consistent 3D models for post-operative analysis, pre-operative planning, and surgical modeling.
    • This method advances the capabilities of intelligent surgery by improving spatial understanding and data-driven analysis.