Viktor Larsson

Viktor Larsson is an Assistant Professor at the Centre for Mathematical Sciences, Lund University, specializing in Computer Vision and Machine Learning. Previously, he worked as a PostDoc and senior researcher at ETH Zurich's Computer Vision and Geometry group under Marc Pollefeys. His research focuses on geometric estimation problems in 3D computer vision, including Structure-from-Motion, visual localization, SLAM, and integrating machine learning into classical pipelines.

Key research interests include robust estimation techniques, hybrid feature usage, and algorithmic efficiency. He actively supervises PhD/MSc students and has secured grants for impactful projects. Recent work explores data-driven alternatives to hand-crafted heuristics in vision tasks.

Notable achievements include a docent qualification (2024), Best Paper Honorable Mention (CVPR 2024), and Best Student Paper (ICCV 2021). His articles span topics like RANSAC optimization, neural scene encoding, and eigenvalue-based localization methods. He also serves on conference committees for CVPR, 3DV, and ECCV, contributing to academic discourse through tutorials and plenary talks.