I am a PhD student at the Computer Vision Group supervised by Prof. Daniel Cremers at the Technical University of Munich.
Before the PhD, I obtained a M.Sc. degree in Systems control and Robotics from KTH, Royal Institute of Technology (Stockholm, Sweden), including an exchange year at ETH Zürich. I also spent roughly 3 years as a computer vision engineer at the Stockholm-based startup Univrses working on visual odometry and deep learning in autonomous driving industry.
I am interested in reliable and high-accuracy 3D perception from in challenging conditions. I approach this problem from various angles using a mix of optimization-based and learning-based methods.
Below is a list of selected publications, for the full list see my Google Scholar.
A probabilistic method combining generative models with measurement consistency for 3D shape reconstruction, traversing Langevin dynamics from a diffusion model while preserving measurement consistency at every step. Full conference version coming soon.
We decouple distortion calibration from 3D reconstruction by working in projective space, averaging multiple pairwise distortion estimates into a single consistent camera model without requiring 3D points or full bundle adjustment.
We build a semantic library (SemLA) of LoRA adapters indexed in CLIP space, from which we can efficiently retrieve a model tailored to each input at test-time.
We improve the robustness and accuracy of surface reconstruction methods by optimizing a symmetric differentiable Chamfer distance for implicit surfaces.
We develop a certifiably optimal method for triangulating noisy observations with outliers.
We develop a globally optimal solution to the noisy hand-eye calibration problem, using a dual quaternion formulation.
Python library for making 3D plots with Blender. Like Matplotlib for 3D visualization, but using Blender's rendering engine. Install with pip install blender-plots.