Arno Solin

Papers

Total Citations

H-Index

About

Arno Solin’s research lies at the intersection of probabilistic machine learning, robotics, and computer vision, with a particular focus on Bayesian nonparametric modeling and sensor fusion. His most influential work addresses the challenge of indoor positioning by modeling ambient magnetic field anomalies through Gaussian processes, leveraging Maxwell’s equations to derive a physically informed, data-efficient interpolation and extrapolation framework. This approach, detailed in his highly cited 2015 and 2018 papers, offers a robust alternative to traditional radio-frequency-based localization, enabling accurate navigation in GPS-denied environments. In the domain of computer vision, Solin has advanced stereo matching generalization through the introduction of visual hints expansion, a technique inspired by the robustness of Visual Inertial Odometry (VIO) that improves depth estimation from sparse, unevenly distributed feature points. His contributions have garnered significant attention, with his magnetic field modeling papers accumulating over a dozen citations and establishing a foundation for subsequent research in indoor positioning. Solin’s work is characterized by a rigorous integration of physical principles with probabilistic methods, making him a notable figure in the development of practical, scalable solutions for autonomous systems and spatial intelligence.