Runze (Patrick) Gan
Research Associate at the Univerisity of Edinburgh
I am currently a Research Associate at the University of Edinburgh. My research primarily focuses on the theoretical and applied aspects of statistical signal processing. More specifically, my research interests include:
- Applications: multi-object tracking, (decentralised) data fusion, manoeuvring object tracking, intent inference, and human–computer interaction.
- Approximate inference methodologies: variational inference, Monte Carlo methods, and other scalable Bayesian techniques.
- Stochastic modelling for time series: covering stochastic differential equations, Lévy processes, and mean-reverting processes.
I obtained my Ph.D. in Information Engineering from the Cambridge University Engineering Department, and previously held research associate roles at the Alan Turing Institute and the University of Cambridge.
My thesis is available here.
My full publication list can be found on Google Scholar.
news
| Jul 16, 2026 | 🎉 The full PiVoT paper is now on arXiv, with complete technical details and a major upgrade for efficiently handling Doppler point-cloud data! Want to see how a real-time, fully training-free joint detector and tracker stacks up against a deep-learning detection benchmark on a full-scale modern automotive radar dataset? Read the paper and explore the automotive radar, maritime surveillance, dense-clutter and thousand-object tracking demos on the PiVoT project page. |
|---|---|
| Jun 17, 2026 | I was delighted to deliver a lecture on multi-target tracking at the Dstl–SPADS Signal Processing Summer School 2026, together with Dr Murat Üney and Prof James Hopgood. |
| Mar 23, 2026 | It was a pleasure to give a talk on PiVoT at the UoE Statistics Seminar, hosted by the School of Mathematics, University of Edinburgh. |
| Apr 08, 2025 | 🚀 Get a sneak peek at our new multi-object tracker, PiVoT! PiVoT is a project I’ve led as the main developer, combining non-trivial technical work with strong empirical performance. It efficiently detects and tracks a large, varying number of objects, estimating their shapes, existence probabilities, and more. PiVoT outperforms existing Poisson measurements-based trackers in accuracy and efficiency across diverse scenes, and notably, enables reliable real-time tracking in highly challenging settings, such as dense clutter or tracking a thousand closely spaced objects in under a second on a standard laptop without gating. 👉 Explore PiVoT and watch demos |
| Apr 03, 2025 | Together with Dr Qing Li and Prof Simon Godsill, I will be co-organising a special session on Advances in Sensor Fusion and Distributed Learning at the 2025 IEEE Statistical Signal Processing Workshop (8–11 June, Edinburgh). More details here. Hope to see you there! |