Ludvig Dillén

PhD Student at Lund University.

prof_pic.jpg

I am a 3rd year PhD student in Computer Vision and Machine Learning at the Centre of Mathematical Sciences at Lund University under the supervision of Viktor Larsson and co-supervision of Magnus Oskarsson. I am funded by ELLIIT and affiliated with WASP.

I research on localization and mapping problems in computer vision, including structure from motion, point cloud registration, structureless and sparse-view localization, as well as scene coordinate regression/reconstruction. In addition to research, I teach undergraduate and graduate courses in mathematics, computer vision, and machine learning, and pursue advanced coursework in these areas to deepen my knowledge.

I am always open to collaborations on projects related to 3D vision, and in particular image-based localization, mapping, and learning good representations. Feel free to reach out!

news

May 2026 Second place in Structured Semantic 3D Reconstruction (S23DR) Challenge 2026 (paper, code)
Feb 2026 Active Exploration for Sparse Visual Localization accepted at CVPR findings track (paper, supp, code)
Feb 2026 Sparse–View Localization via Online Neural 3D Regression accepted at CVPR (paper, supp, poster, video, code)
Mar 2025 FACT: Multinomial Misalignment Classification for Point Cloud Registration accepted at SCIA (paper+supp, code)
Oct 2023 Finished Civil Engineering at Linköping University with MSc in Computer Vision and Signal Analysis with GPA 4.92/5.00 and Feature-Aware Point Transformer for Point Cloud Alignment Classification