Scholar
Paul-Edouard Sarlin
Google Scholar ID: 2fAUgfAAAAAJ
Researcher at Google
Computer Vision
Deep Learning
3D Geometry
Visual Localization
Structure-from-Motion
Follow
Homepage
↗
Google Scholar
↗
Citations & Impact
All-time
Citations
6,679
H-index
12
i10-index
12
Publications
18
Co-authors
30
list available
Contact
Email
paul.edouard.sarlin@gmail.com
CV
Open ↗
Twitter
Open ↗
GitHub
Open ↗
LinkedIn
Open ↗
Publications
6 items
VidMap: Exploiting Temporal Structure for Video-Based Structure-from-Motion
2026
Cited
0
UNIGEOCLIP: Unified Geospatial Contrastive Learning
2026
Cited
0
RaCo: Ranking and Covariance for Practical Learned Keypoints
2026
Cited
0
Scaling Image Geo-Localization to Continent Level
2025
Cited
0
Benchmarking Egocentric Visual-Inertial SLAM at City Scale
2025
Cited
0
MP-SfM: Monocular Surface Priors for Robust Structure-from-Motion
2025
Cited
0
Resume
Academic Achievements
2025: Released a new dataset and benchmark for large-scale egocentric SLAM; paper accepted to ICCV 2025
2025: Received DAGM MVTec Dissertation Award at GCPR 2025
2025: Presented MP-SfM at CVPR 2025
2024: Awarded the prestigious ETH Silver Medal for PhD thesis
2024: Presented GeoCalib at ECCV 2024
2023: LightGlue accepted to ICCV 2023
2023: Worked on SNAP framework during Google internship
2023: OrienterNet accepted to CVPR 2023
2022: Released LaMAR benchmark for AR-based visual localization (ECCV 2022)
2022: Received Best Presentation Award at ICVSS 2022
2021: Pixel-Perfect Structure-from-Motion received Best Student Paper Award at ICCV 2021
2021: Back to the Feature published at CVPR 2021
2020: SuperGlue and hloc won 4 ECCV 2020 visual localization challenges
2020: SuperGlue won 3 CVPR 2020 competitions and was selected for oral presentation
2019: Hierarchical Localization won the Visual Localization Challenge at CVPR 2019 workshop
2019: Fishyscapes Benchmark presented orally at ICCV 2019 workshop on Autonomous Driving
Research Experience
Researcher at Google Geo (Maps, StreetView, Earth), Zurich (since Oct 2024): working on mapping problems
Intern at Google Research Zurich: developed SNAP, a self-supervised framework for neural map inference from streetview and aerial images
Research Intern at Facebook Reality Labs, London: worked on OrienterNet
Research Intern at Microsoft Mixed Reality & AI Lab, Zurich
Conducted Master’s thesis at Magic Leap, San Francisco
Spent time at ASL, Traverse.ai, NUS, and EPFL
Co-authors
13 total
Marc Pollefeys
Professor of Computer Science, ETH Zurich, and Director Spatial AI Lab, Microsoft
Daniel DeTone
Research Scientist at Meta Reality Labs
Cesar Cadena
Senior Scientist (tenured), ETH Zurich
Roland Siegwart
Professor of Robotics, ETH Zurich, Switzerland
Viktor Larsson
Assistant Professor at Lund University
Fredrik Kahl
Professor, Chalmers University of Technology
Vincent Lepetit
Ecole des Ponts ParisTech
Torsten Sattler
Senior Researcher, Czech Technical University in Prague