Javier Morlana

Co-founder at EndoCartoScope · Universidad de Zaragoza

Building Spatial AI for endoscopy.

PhD in SLAM and visual place recognition for colonoscopy at the Robotics Lab, under J.M.M. Montiel. Research internships at Google and Leica Geosystems, both on visual-inertial SLAM.

CV
Javier Morlana

Now

EIC-funded · 2025-2028

EndoCartoScope

Spatial AI for medical endoscopy: we build a per-patient 3D map of the cavity while localizing the endoscope inside it, using only the video of a standard endoscope, enough to tell how much of the mucosa was inspected, guide navigation and measure lesions.

From TRL4 to TRL6, on the way to a spin-off.

Project site

News

  • Sep 2026

    Organizing CLiMB at MICCAI 2026

    The Colonoscopy Localization and Mapping Benchmark, a sub-challenge of EndoVis. Challenge day 27 September, Strasbourg.

    Challenge siteEndoVis

  • Sep 2026

    Organizing DCA-MI at ECCV 2026

    Second edition of the workshop on data curation and augmentation in medical imaging. 9 September, Malmö.

    Workshop site

  • Jul 2025

    EndoCartoScope starts

    EIC-funded project at Universidad de Zaragoza. I join as co-founder.

    Project site

  • Jun 2025

    PhD defended

    SLAM and visual place recognition for colonoscopy, Universidad de Zaragoza.

Publications

ColonSLAM: a metric SLAM produces a linear graph of 3D submaps, which becomes a graph with covisibility relationships.

Topological SLAM in colonoscopies leveraging deep features and topological priors

Javier Morlana, Juan D. Tardós, J.M.M. Montiel

MICCAI 2024

Paper Code Publisher

ColonMapper: a topological map of the colon is built from one sequence, then a second sequence is localized against it.

ColonMapper: topological mapping and localization for colonoscopy

Javier Morlana, Juan D. Tardós, J.M.M. Montiel

ICRA 2024

Paper Code Talk Publisher

DRAN: a single network producing global descriptors, hierarchical features for re-ranking and dense features for alignment.

Reuse your features: unifying retrieval and feature-metric alignment

Javier Morlana, J.M.M. Montiel

ICRA 2023

Paper Code Talk Publisher

Self-supervised Visual Place Recognition for Colonoscopy Sequences

Javier Morlana, Pablo Azagra, Javier Civera, J.M.M. Montiel

MIDL 2021 · short paper

Paper Talk Publisher

SD-DefSLAM: semi-direct monocular SLAM tracking a deforming intracorporeal scene.

SD-DefSLAM: Semi-Direct Monocular SLAM for Deformable and Intracorporeal Scenes

Juan J. Gómez-Rodríguez, José Lamarca, Javier Morlana, Juan D. Tardós, José M.M. Montiel

ICRA 2021

Paper Code Publisher

Experience

2025-present

Co-founder

EndoCartoScope · Universidad de Zaragoza

Zaragoza, Spain

2020-2025

PhD in Computer Vision

Universidad de Zaragoza

SLAM and visual place recognition for colonoscopy, advised by J.M.M. Montiel. Defended June 2025.

Sep-Dec 2022

Student Researcher

Google · Zurich, Switzerland

SLAM aided by neural-network priors, in Federico Tombari's team.

Aug 2019-Feb 2020

Visual-Inertial SLAM intern

Leica Geosystems · Switzerland

Built the Leica-UZ visual-inertial SLAM library, combining VI-ORB-SLAM and VINS-Mono.

Oct 2017-Jun 2018

Research grant

Universidad de Zaragoza

Extended ORB-SLAM to non-rigid scenes.