Ph.D. Student · Computer Vision and Robotics
Technical University of Munich
I am a Ph.D. student at the Technical University of Munich, sponsored by Siemens AG and supervised by Prof. Slobodan Ilic. My doctoral research focuses on vision-language-action (VLA) models for robot manipulation.
Previously, I completed my M.Sc. in Robotics, Cognition, and Intelligence at the Technical University of Munich. My research focuses on 3D computer vision, with a specific interest in 6D pose estimation, object detection, and 3D scene understanding.
Before my time at TUM, I earned my B.Eng. in Mechatronics from Tongji University.
European Conference on Computer Vision (ECCV), 2026.
PANY is a model-free 6D object pose estimation framework that works from arbitrary RGB or RGB-D reference views and generalizes to novel objects.
arXiv / PDF / Project Page / Code
European Conference on Computer Vision (ECCV), 2026.
XYZ-IBD is a benchmark for object detection and 6D pose estimation in industrial bin-picking scenes with metallic, specular, and highly occluded objects.
arXiv / PDF / Project Page / Code
British Machine Vision Conference (BMVC), 2026.
SOS segments unseen target objects from a single reference image using an object-conditional transformer, without CAD models or template onboarding.
arXiv / PDF / Project Page / Code
ICCV 2025 Workshop on Recovering 6D Object Pose, 2025.
BioDet improves industrial object detection for downstream 6D pose estimation through low-light enhancement and background removal with foundation models.