About
Welcome! I am a doctoral student of Computer Science in the Computer Vision and Geometry Group at ETH Zürich, where I am advised by Marc Pollefeys (ETH) and co-advised by Jan-Michael Frahm (UNC).
I received my Bachelor of Science from TU Munich and my Master of Science from UNC Chapel Hill. I worked in the computer vision group at UNC, and in the Remote Sensing Technology and the Photogrammetry and Remote Sensing departments at TUM. I also spent time at Microsoft Research, Google, and the German Aerospace Center.
Research
My main research interests lie in robust image-based 3D modeling. More broadly, I am interested in computer vision, geometry, structure-from-motion, (multi-view) stereo, localization, optimization, machine learning, and image processing. You can find my publications on Google Scholar.
Open Source
As part of my research, I developed the open-source software COLMAP - an end-to-end image-based 3D reconstruction software, which achieves state-of-the-art results on recent benchmarks (1, 2). Over the years, I also contributed to a number of scientific open source projects, including scikit-image, scikit-learn, NumPy, SciPy, Ceres-Solver, etc. You can find my open source activities on GitHub.
Publications [Google Scholar]
2018
-
Semantic Visual Localization Conference on Computer Vision and Pattern Recognition (CVPR)
Paper / Supplementary / Bibtex
-
Augmenting Crowd-Sourced 3D Reconstructions
using Semantic Detections Conference on Computer Vision and Pattern Recognition (CVPR)Paper / Supplementary / Bibtex
2017
-
Comparative Evaluation of Hand-Crafted and Learned Local Features Conference on Computer Vision and Pattern Recognition (CVPR)
Benchmark / Paper / Supplementary / Bibtex
-
A Multi-View Stereo Benchmark with High-Resolution Images and Multi-Camera Videos Conference on Computer Vision and Pattern Recognition (CVPR)
Benchmark / Paper / Supplementary / Bibtex
-
Tutorial: Large-scale 3D Modeling from Crowdsourced Data Conference on Computer Vision and Pattern Recognition (CVPR)
Mapping on the Fly: Real-Time 3D Dense Reconstruction, Digital Surface Map and Incremental Orthomosaic Generation for Unmanned Aerial Vehicles International Conference on Field and Service Robotics (FSR)
Code / Paper / Supplementary / Bibtex
2016
-
A Vote-and-Verify Strategy
for Fast Spatial Verification in Image Retrieval Asian Conference on Computer Vision (ACCV)Code / Paper / Supplementary / Bibtex
-
Tutorial: Large-scale 3D Modeling from Crowdsourced Data International Conference on 3D Vision (3DV)
-
Pixelwise View Selection for Unstructured Multi-View Stereo European Conference on Computer Vision (ECCV)
Paper / Supplementary / Bibtex / Video
-
Indoor-Outdoor 3D Reconstruction Alignment European Conference on Computer Vision (ECCV)
Paper / Supplementary / Bibtex / Video
-
Structure-from-Motion Revisited Conference on Computer Vision and Pattern Recognition (CVPR)
Website / Paper / Supplementary / Bibtex / Video
-
From Dusk till Dawn: Modeling in the Dark Conference on Computer Vision and Pattern Recognition (CVPR)
Paper / Supplementary / Bibtex / Video
2015
-
Efficient Two-View Geometry Classification German Conference on Pattern Recognition (GCPR) Best Paper Honorable Mention
-
PAIGE: PAirwise Image Geometry Encoding
for Improved Efficiency in Structure-from-Motion Conference on Computer Vision and Pattern Recognition (CVPR) -
From Single Image Query to Detailed 3D Reconstruction Conference on Computer Vision and Pattern Recognition (CVPR)
-
Reconstructing the World* in Six Days
*(As Captured by the Yahoo 100 Million Image Dataset) Conference on Computer Vision and Pattern Recognition (CVPR)
2014
-
scikit-image: Image processing in Python PeerJ
-
Structure-from-motion for MAV image sequence analysis
with photogrammetric applications Photogrammetric Computer Vision Symposium (PCV) -
Benefit of Using Multiple Baselines and Multiple Aspects
for SAR Interferometry of Urban Areas IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing (JSTARS) -
Adaptive Covariance Matrix Estimation
for Multi-Baseline InSAR Data Stacks IEEE Transactions on Geoscience and Remote Sensing (TGARS)