I am currently a 4th-year PhD student in the VLC (Vision, Learning and Control) group, University of Southampton, supervised by Xiaohao Cai. My study and research are mainly focused on 3D object detection problems with images and point cloud data. I obtained my Master’s degree at the University of Southampton in 2021. Before that, I obtained my bachelor’s degree in Mathematics in 2020.
My research interest includes: 3D Computer Vision, Cross-domain 3D Object Detection, 3D Visual Grounding.
🎉 I will start a new role as a Research Fellow in Deep Learning for Robot Audition at the University of Southampton from September 2025.
🔥 News
- 2025.01: 🎉🎉 Great news! Our paper Talk2Radar has been accepted by ICRA 2025.
- 2024.08: 🎉🎉 One paper has been accepted by ICONIP 2024.
- 2024.07: 🎉🎉 One paper has been accepted by ECAI 2024.
📝 Publications
🚗 3D Object Detection

CornerPoint3D: Look at the Nearest Corner Instead of the Center
Ruixiao Zhang, Runwei Guan, Xiangyu Chen, Adam Prugel-Bennett, Xiaohao Cai
We propose a novel 3D object detector named CornerPoint3D, which is built upon CenterPoint and uses heatmaps to supervise the learning and detection of the nearest corner of each object. CornerPoint3D realizes a balanced trade-off between the detection quality of entire bounding boxes and the locating accuracy of closer surfaces to the LiDAR sensor, providing a more practically reasonable and robust cross-domain 3D object detection solution.

Detect Closer Surfaces That Can be Seen: New Modeling and Evaluation in Cross-Domain 3D Object Detection
Ruixiao Zhang, Yihong Wu, Juheon Lee, Xiaohao Cai, Adam Prugel-Bennett
Considering the properties of LiDAR point cloud, we propose a refinement head named EdgeHead, with two additional evaluation metrics, the CS-BEV and CS-ABS, to focus more on the closer surfaces of detected objects instead of their entire boxes.

Revisiting Cross-Domain Problem for LiDAR-based 3D Object Detection
Ruixiao Zhang, Juheon Lee, Xiaohao Cai, Adam Prugel-Bennett
🔍 3D Visual Grounding

Talk2Radar: Bridging Natural Language with 4D mmWave Radar for 3D Referring Expression Comprehension
Runwei Guan, Ruixiao Zhang, Ningwei Ouyang, Jianan Liu, Ka Lok Man, Xiaohao Cai, Ming Xu, Jeremy Smith, Eng Gee Lim, Yutao Yue, Hui Xiong
Others
The Journal of Supercomputing MAN and CAT: mix attention to nn and concatenate attention to YOLO, Runwei Guan, Ka Lok Man, Haocheng Zhao, Ruixiao Zhang, Shanliang Yao, Jeremy Smith, Eng Gee Lim, Yutao Yue
ECAI 2023 A Bilevel Formalism for the Peer-Reviewing Problem, Gennaro Auricchio, Ruixiao Zhang, Jie Zhang, Xiaohao Cai
📃 Reviewer Experience
- European Conference on Artificial Intelligence (ECAI)
- International Conference on Neural Information Processing (ICONIP)
📖 Educations
- 2021.09 - 2025.06 (now), PhD in Computer Science, University of Southampton, UK.
- 2020.09 - 2021.09, Master of Science in Data Science, University of Southampton, UK.
- 2016.09 - 2020.07, Undergraduate in Mathematics and Applied Mathematics, East China Normal University, China.
💻 Internships
- 2019.07 - 2019.09, Data Analyst Intern. Squirrel AI, Shanghai, China.