Summary
Ming Xu is a postdoctoral researcher with ~8 years of experience at the intersection of robotics and computer vision, currently working at EPFL's CVLab and previously at ANU on industry-collaborative projects with Seeing Machines. He combines principled mathematical modeling with hands-on algorithm design for constrained optimisation problems, covering optimal control, self-supervised segmentation, point-set registration, mesh optimisation and local feature matching. His work has been validated in challenging real-world settings and his unsupervised long-form video segmentation paper was nominated for a CVPR 2024 Best Paper Award. Ming has a strong quantitative background (PhD in Mechatronics/Robotics, MS in Mathematics, actuarial undergrad) and experience translating research into deployed systems, including AV pose-estimation pipelines developed with Ford collaborators. He also co-supervises PhD students across anomaly detection, foundation models and generative prediction, and is now applying neural-network-based CAD design to fixed-wing UAVs under an SNSF Advanced Grant. Colleagues describe him as a researcher who pairs rigorous theory with pragmatic engineering to push academic ideas toward industry impact.
8 years of coding experience
1 year of employment as a software developer
Doctor of Philosophy - PhD, Mechatronics, Robotics, and Automation Engineering, Doctor of Philosophy - PhD, Mechatronics, Robotics, and Automation Engineering at QUT (Queensland University of Technology)
Victorian Certificate of Education (VCE), Victorian Certificate of Education (VCE) at Melbourne High School
The University of Melbourne
UNSW Sydney
English, Chinese