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

Multi-modal monocular endoscopic depth and pose estimation with edge-guided self-supervision.

Xinwei Ju1, Rema Daher2, Danail Stoyanov2

  • 1Department of Computer Science, Hawkes Institute, University College London, 43-45 Foley Street, London, W1W 7TY, UK. xinwei.ju.22@ucl.ac.uk.

International Journal of Computer Assisted Radiology and Surgery
|May 10, 2026
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

Depth Perception and Spatial Vision01:15

Depth Perception and Spatial Vision

Depth perception is the ability to perceive objects three-dimensionally. It relies on two types of cues: binocular and monocular. Binocular cues depend on the combination of images from both eyes and how the eyes work together. Since the eyes are in slightly different positions, each eye captures a slightly different image. This disparity between images, known as binocular disparity, helps the brain interpret depth. When the brain compares these images, it determines the distance to an object.

You might also read

Related Articles

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

Sort by
Same author

EndoLRMGS: Combining Large Reconstruction Modelling and Gaussian Splatting for Complete Endoscopic Scene Reconstruction.

IEEE transactions on medical imaging·2026
Same author

Robust quantification of ICG fluorescence perfusion in neonatal bowel surgery via deep point tracking.

International journal of computer assisted radiology and surgery·2026
Same author

IJCARS: IPCAI 2026 special issue-17th international conference on information processing in computer-assisted interventions 2026.

International journal of computer assisted radiology and surgery·2026
Same author

Blob representation of robotic surgical scenes for position-aware video generation.

International journal of computer assisted radiology and surgery·2026
Same author

VerTE-MT: A Multi-Task Framework with Entropy-Guided Sampling for Vertebrae Segmentation and Localisation in CT.

IEEE journal of biomedical and health informatics·2026
Same author

SurgViVQA: temporally grounded video question answering for surgical scene understanding.

International journal of computer assisted radiology and surgery·2026

This study introduces a self-supervised learning framework for monocular depth and pose estimation in colonoscopy. The method improves geometric understanding for safer, marker-free colonoscopy navigation by using edge maps and luminance decomposition.

Area of Science:

  • Medical Imaging
  • Computer Vision
  • Robotics

Background:

  • Monocular depth and pose estimation are crucial for colonoscopy navigation.
  • Challenges include texture-less surfaces, complex lighting, and tissue deformation.
  • Lack of in-vivo datasets with ground truth hinders development.

Purpose of the Study:

  • To develop a self-supervised learning framework for accurate monocular depth and pose estimation in colonoscopy.
  • To enhance geometric understanding for improved colonoscopy-assisted navigation systems.
  • To address the challenges of texture-less surfaces, complex illumination, and tissue deformation.

Main Methods:

  • A self-supervised learning framework using anatomical and illumination priors.
  • Integration of edge maps from a learning-based detector for mucosal boundaries.
Keywords:
Colonoscopy navigationMonocular depth estimationPose estimationSelf-supervised learning

Related Experiment Videos

  • Luminance decomposition via intrinsic image separation to isolate shading from reflectance.
  • Edge-guided loss for stage-wise refinement, enhancing motion alignment and depth consistency.

    • Main Results:

    • State-of-the-art performance in depth estimation and competitive accuracy in pose estimation on phantom and real datasets.
    • Self-supervised training on real data surpassed supervised training on phantom data.
    • Dataset-specific frame-rate sampling is critical for effective training sequences.

    Conclusions:

    • The proposed framework enhances geometric learning in endoscopic videos using structure- and illumination-aware cues.
    • Provides a robust foundation for reliable, marker-free colonoscopy navigation.
    • Code and pretrained models are publicly available.