Unsupervised Odometry and Depth Learning for Endoscopic Capsule Robots

TL;DR

This paper presents a fully unsupervised, real-time odometry and depth estimation framework for monocular endoscopic capsule robots, enhancing minimally invasive diagnostics and interventions in the gastrointestinal tract.

Contribution

It introduces a novel unsupervised learning approach combining view warping and re-projection loss for simultaneous odometry and depth estimation in endoscopic capsules.

Findings

Effective motion estimation demonstrated on porcine stomach datasets.

Accurate depth recovery validated through quantitative and qualitative analyses.

Real-time processing capability achieved.

Abstract

In the last decade, many medical companies and research groups have tried to convert passive capsule endoscopes as an emerging and minimally invasive diagnostic technology into actively steerable endoscopic capsule robots which will provide more intuitive disease detection, targeted drug delivery and biopsy-like operations in the gastrointestinal(GI) tract. In this study, we introduce a fully unsupervised, real-time odometry and depth learner for monocular endoscopic capsule robots. We establish the supervision by warping view sequences and assigning the re-projection minimization to the loss function, which we adopt in multi-view pose estimation and single-view depth estimation network. Detailed quantitative and qualitative analyses of the proposed framework performed on non-rigidly deformable ex-vivo porcine stomach datasets proves the effectiveness of the method in terms of motion estimation and depth recovery.