Chris Xiaoxuan Lu
Associate Professor @ Department of Computer Science, University College London
I am an Associate Professor in Robotics and AI at the Department of Computer Science, University College London (UCL). I delve into the dynamic field of spatial AI. My expertise encompasses developing autonomous vehicles and mixed/augmented reality (MR/AR) and IoT devices using AI methods, with a keen focus on elevating their spatial perception ability in 3D environments. My research unfolds across three pivotal dimensions:
- Pose Estimation – Determining the orientation and position of moving agents in the wild.
- Mapping and Reconstruction – Creating generalisable and expressive representations of environments.
- Scene Understanding – Interpreting and making sense of complex and dynamic scenes on-the-fly.
At the core of my research is tackling real-world robustness challenges imposed on spatial perception. This includes overcoming hurdles like visual degradation, GPS denial, navigating the limitations of available resources, and handling unpredictable, out-of-distribution samples. Discover more about my research projects on my research page.
Before coming to UCL, I was a faculty member at the University of Edinburgh and University of Liverpool. I did both my PhD study and post-doctoral in the Department of Computer Science, University of Oxford. Even earlier, I received my M.Eng degree from Nanyang Technological University (NTU).
news
Sep 26, 2024 | 🎉 One papers accepted to NeurIPS-2024. Stay tuned for the code and video release. See you in Vancouver, Canada. |
---|---|
Jul 2, 2024 | 🎉 Two papers accepted to ECCV-2024 and one paper accepted to IROS-2024. Stay tuned for the code and video release. See you at Milano and Abu Dhabi. |
Jan 29, 2024 | 🎉 Three papers accepted to ICRA-2024. Stay tuned for the code and video release. See you at Yokohama in May. |
selected publications
- NeurIPS’24RadarOcc: Robust 3D Occupancy Prediction with 4D Imaging RadarIn The Thirty-eighth Conference on Neural Information Processing Systems 2024