Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

EndoMetric: Near-Light Monocular Metric Scale Estimation in Endoscopy

Raúl Iranzo, Víctor M. Batlle, Juan D. Tardós, José M. M. Montiel

TL;DR

Monocular endoscopes produce reconstructions with unknown metric scale, hindering precise measurements. We introduce EndoMetric, a fully model-based method that first obtains an up-to-scale 3D structure and then estimates the true metric scale by fitting a near-light photometric model that accounts for the light-camera baseline and inverse-square falloff, jointly solving for scale, albedo, and gains. Key contributions include a scale-dependent near-light photometric model, a Levenberg–Marquardt optimization for scale with robust costs, and an initialization strategy; validated in simulations and the EndoMapper dataset showing sub-centimeter polyp measurements. The results demonstrate metric-scale endoscopy without hardware modification or task-specific learning, enabling accurate measurements and opening pathways toward true-scale SLAM and autonomous endoscopic interventions.

Abstract

Geometric reconstruction and SLAM with endoscopic images have advanced significantly in recent years. In most medical fields, monocular endoscopes are employed, and the algorithms used are typically adaptations of those designed for external environments, resulting in 3D reconstructions with an unknown scale factor. For the first time, we propose a method to estimate the real metric scale of a 3D reconstruction from standard monocular endoscopic images without relying on application-specific learned priors. Our fully model-based approach leverages the near-light sources embedded in endoscopes, positioned at a small but nonzero baseline from the camera, in combination with the inverse-square law of light attenuation, to accurately recover the metric scale from scratch. This enables the transformation of any endoscope into a metric device, which is crucial for applications such as measuring polyps, stenosis, or assessing the extent of diseased tissue.

EndoMetric: Near-Light Monocular Metric Scale Estimation in Endoscopy

TL;DR

Monocular endoscopes produce reconstructions with unknown metric scale, hindering precise measurements. We introduce EndoMetric, a fully model-based method that first obtains an up-to-scale 3D structure and then estimates the true metric scale by fitting a near-light photometric model that accounts for the light-camera baseline and inverse-square falloff, jointly solving for scale, albedo, and gains. Key contributions include a scale-dependent near-light photometric model, a Levenberg–Marquardt optimization for scale with robust costs, and an initialization strategy; validated in simulations and the EndoMapper dataset showing sub-centimeter polyp measurements. The results demonstrate metric-scale endoscopy without hardware modification or task-specific learning, enabling accurate measurements and opening pathways toward true-scale SLAM and autonomous endoscopic interventions.

Abstract

Geometric reconstruction and SLAM with endoscopic images have advanced significantly in recent years. In most medical fields, monocular endoscopes are employed, and the algorithms used are typically adaptations of those designed for external environments, resulting in 3D reconstructions with an unknown scale factor. For the first time, we propose a method to estimate the real metric scale of a 3D reconstruction from standard monocular endoscopic images without relying on application-specific learned priors. Our fully model-based approach leverages the near-light sources embedded in endoscopes, positioned at a small but nonzero baseline from the camera, in combination with the inverse-square law of light attenuation, to accurately recover the metric scale from scratch. This enables the transformation of any endoscope into a metric device, which is crucial for applications such as measuring polyps, stenosis, or assessing the extent of diseased tissue.

Paper Structure

This paper contains 19 sections, 6 equations, 3 figures, 1 table.

Table of Contents

  1. Introduction
  2. Related work
  3. Fundamentals
  4. EndoMetric
  5. Up-to-scale Multi-view Reconstruction
  6. Near-light Photometric Model
  7. Metric Scale Estimation
  8. Initial Guess for the Scale
  9. Experiments
  10. Datasets
  11. Simulation dataset.
  12. EndoMapper dataset azagra2023endomapper.
  13. Impact of Distance to the Surface
  14. Impact of Multi-View Reconstruction Accuracy
  15. Impact of Initial Guess
  16. ...and 4 more sections

Figures (3)

  • Figure 1: Our method estimates the metric scale factor $\lambda$ by leveraging a near-light illumination model applied to multi-view images captured with a monocular endoscope.
  • Figure 2: Method accuracy depends on surface distance, performing best when it matches the light-camera baseline. Left: Scale error increases from 1% to 5% with distance. Right: Greater distances weaken the photometric cost function’s minimum.
  • Figure 3: Results in EndoMapper dataset azagra2023endomapper.