Rong Zou
Ph.D. Student in Robotic Perception
Robotics and Perception Group, University of Zurich / ETH Zurich / ETH AI Center
Research Interests
Beyond Research
I'm happiest in forests, on mountain trails, or anywhere that feels a little wild. If we ever cross paths, it might be on a shaded trail somewhere in the Alps.
I love capturing small moments of the world with my camera, chasing strange ideas, and getting lost in imaginative stories. I used to read a lot of scifi, and Liu Cixin's The Three-Body Problem is still one of my favorites.
I also have a soft spot for strategy games, especially Jieqi (dark Chinese chess) and Yingxiongsha (a variant of Legends of the Three Kingdoms), where the fun lies in reading people, timing the right move, and surviving beautifully bad situations.
About
I am pursuing a Ph.D. in robotic perception with the Robotics and Perception Group, led by Prof. Davide Scaramuzza. I am also an associated doctoral researcher with the ETH AI Center.
My research interests sit at the intersection of vision, learning, and robotics, with a focus on event-based visual sensing and transferable representation learning for robotic tasks such as motion estimation, scene understanding, and robust perception in challenging conditions.
Before starting my Ph.D., I completed my Master's degree with distinction in Robotics, Systems, and Control at ETH Zurich. I also worked at the Computer Vision and Geometry Group on object retrieval under motion blur, and at Huawei Switzerland on image quality enhancement.
Prior to my Master's, I received my Bachelor's degree with distinction in Engineering from Huazhong University of Science and Technology.
Selected Publications
View All ->Event-Aided Sharp Radiance Field Reconstruction for Fast-Flying Drones
Rong Zou*, Marco Cannici*, Davide Scaramuzza
IEEE Transactions on Robotics (TRO), 2026
A unified event-and-frame framework for sharp radiance field reconstruction from fast drone flights affected by motion blur and noisy pose estimates.
Retrieval Robust to Object Motion Blur
Rong Zou, Marc Pollefeys, Denys Rozumnyi
European Conference on Computer Vision (ECCV), 2024
A method and dataset for learning robust representations capable of bidirectional matching between motion-blurred objects and their deblurred counterparts.
Seeing Behind Dynamic Occlusions with Event Cameras
Rong Zou, Manasi Muglikar, Nico Messikommer, Davide Scaramuzza
arXiv preprint, 2023
A data-driven approach to reconstruct background appearance from a single viewpoint in the presence of dynamic occlusions.
News
Our paper "Event-Aided Sharp Radiance Field Reconstruction for Fast-Flying Drones" appeared in IEEE Transactions on Robotics.
Our paper "Low-latency Event-based Object Detection with Spatially-Sparse Linear Attention" is available on arXiv.
Our paper "FastEventDGS: Deformable Gaussian Splatting for Fast Dynamic Scenes from a Single Event Camera" was accepted by CVPR 2026.
Our paper "Event-Aided Sharp Radiance Field Reconstruction for Fast-Flying Drones" was submitted to IEEE Transactions on Robotics.
Started PhD at the Robotics and Perception Group, Institute of Neuroinformatics, UZH and ETH Zurich and joined ETH AI Center as an associated doctoral student.