I am a Ph.D. student and full-time researcher. My supervisor is Prof. Anne Driemel. My current projects revolve around similarity measure between curves.

My primary research interest is the design and engineering of algorithms and data structures for computational and geometric problems in low-dimensional spaces. Whenever I am not thinking about theoretical results, I enjoy verifying the merit of my theoretical work by implementing and carrying out experiments on real-world data.

Visit my website.

Througout my Ph.D. studies I wrote the following pieces of software.

Clustering trajectories is a central challenge when confronted with large amounts of movement data such as GPS data. We study a clustering problem that can be stated as a geometric set cover problem: Given a polygonal curve, find a small set of low-complexity representative trajectories such that any point on the input trajectories lies on a subtrajectory of the input that has small Fréchet distance to one of the representative trajectories.

We ran extensive experiments on GPS-data and full-body motion data with this software and found that its efficient running time makes our approach viable and produces high quality outputs.

The underlying approach is a collaboration with Anne Driemel. In the later stages Paul Jünger and Simon Bartlmae joined in. The github repository can be found here.

The following problem was posed as the CG:SHOP2025 challenge:

In the Planar Straight Line Graphs (PSLGs) variant of the Minimum Non-Obtuse Triangulation problem, the objective is to triangulate the area of a given PSLG G, which is defined by a set of vertices and edges laid out in a plane such that the edges do not cross except at their endpoints. The triangulation must incorporate the existing edges of G and can include the addition of Steiner points anywhere in the plane, including on these edges. The placement of Steiner points is particularly challenging because it can affect the geometric properties of adjacent faces, thus complicating the triangulation process. All triangles formed in the solution must be non-obtuse, with each angle not exceeding 90 degrees, and the solution seeks to minimize the total number of Steiner points.

My team consisting of Mikkel Abrahamsen, Florestan Brunck, Benedikt Kolbe, André Nusser and me placed first and were invited to present our algorithmic approach at SoCG 2025. The github repository can be found here.