Ruiyang Hao

Ruiyang Hao

I am currently a Ph.D. student in the Translational AI Research Lab at King’s College London (KCL), advised by Dr. Yunpeng Li and Prof. Yulan He. My research focuses on multimodal artificial intelligence for dental diagnosis, with the goal of advancing healthcare applications through cutting-edge machine learning methods.

Before joining KCL, I worked as a Research Assistant at the Institute for AI Industry Research (AIR), Tsinghua University, where I collaborated closely with Prof. Zaiqing Nie and Dr. Haibao Yu on research related to autonomous driving perception, planning and simulation. I obtained my Master’s degree from the Department of Automation, Tsinghua University, supervised by Prof. Biqing Huang. My master’s research primarily focused on computer vision for industrial defect inspection. I received my Bachelor degree (with Hons.) from Shenyuan Honors College, Beihang University.

I have published several papers in high-level AI conferences and journals, with total citations on Google Scholar.

My current research interests lie in multimodal learning, computer vision, and medical AI, with a special focus on their applications in dentistry and healthcare.

📝 Publications

arXiv 2025
sym

StyleDrive: Towards Driving-Style Aware Benchmarking of End-To-End Autonomous Driving

Ruiyang Hao, Bowen Jing, Haibao Yu, Zaiqing Nie

arXiv: 2506.23982, Under Review. Project, Ref

  • Novel large-scale real-world personalized driving dataset
  • High-quality multi-stage annotation pipeline
  • First standardized benchmark for personalized E2EAD
  • Significant gains in human-like behavior via personalization
arXiv 2025
sym

DriveE2E: Closed-Loop Benchmark for End-to-End Autonomous Driving through Real-to-Simulation

Haibao Yu*, Wenxian Yang*, Ruiyang Hao*, Chuanye Wang*, Jiaru Zhong*, Ping Luo, Zaiqing Nie (* means equal contributions)

arXiv: 2509.23922, Under Review. Project, Ref

  • Infra-view-enhanced real2sim framework for realistic closed-loop evaluation.
  • 15 high-fidelity digital twins & 800 infra-sensor diverse scenarios.
  • Comprehensive closed-loop benchmark with expert dataset for SOTA evaluation and IL training.
CVPR 2024
sym

RCooper: A Real-world Large-scale Dataset for Roadside Cooperative Perception

Ruiyang Hao, Siqi Fan, Yingru Dai, Zhenlin Zhang, Chenxi Li, Yuntian Wang, Haibao Yu, Wenxian Yang, Jirui Yuan, Zaiqing Nie

IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2024), 22347-22357. Project, Ref

  • 1st real-world large-scale dataset for roadside cooperative perception.
  • 50k images & 30k point clouds with 3D annotations across 10 semantic classes.
  • Benchmarks on 3D detection & tracking with SOTA results, proving effectiveness of cooperation.
AEI 2022
sym

Efficient surface defect detection using self-supervised learning strategy and segmentation network

Ronge Xu*, Ruiyang Hao*, Biqing Huang (* means equal contributions)

Advanced Engineering Informatics, 2022, 52: 101566. Ref

  • Lightweight defect detection for efficient and accurate industrial inspection.
  • Attention & multi-task strategy to enhance defect focus and edge segmentation.
  • Self-supervised learning with homographic enhancement and synthetic datasets to reduce annotation cost and improve generalization.
JIM 2021
sym

A steel surface defect inspection approach towards smart industrial monitoring

Ruiyang Hao, Bingyu Lu, Ying Cheng, Xiu Li, Biqing Huang

Journal of Intelligent Manufacturing, 2021, 32(7): 1833-1843. Ref

  • DIN: steel defect inspection network balancing efficiency and performance.
  • Deformable convolution backbone for adaptive feature extraction on diverse defect shapes.
  • Balanced feature pyramid fusion for accurate multi-size defect inspection.
CAIE 2021
sym

Manufacturing service supply-demand optimization with dual diversities for industrial internet platforms

Ruiyang Hao, Ying Cheng, Yongping Zhang, Fei Tao

Computers & Industrial Engineering, 2021, 156: 107237. Ref

  • Classified hypernetwork & MS-SDR for modeling dual diversities in supply-demand.
  • MS-SDO model with stakeholder analysis to enhance platform stability and utility.
  • Knowledge-guided NSGAII + experiments for effective optimization and management insights.

🎖 Honors and Awards

  • 2025.10 Newland Pedley Studentship, awarded by King’s College London.
  • 2018.12 National Scholarship, the Top Scholarship in China, top 1%.
  • 2022.10 First-class Excellence Scholarship, awarded by Tsinghua University, top 2.5%.
  • 2021.10 First-class Excellence Scholarship, awarded by Tsinghua University, top 2.5%.
  • 2020.10 Outstanding graduate & the honored degree, awarded by Beihang University, top 5%.
  • 2019.4 Meritorious Prize of Interdisciplinary Contest in Modeling, awarded by COMAP, top 8% of contestants.
  • 2018.4 Meritorious Prize of Interdisciplinary Contest in Modeling, awarded by COMAP, top 8% of contestants.

📖 Educations

  • 2025.10 - now, PhD, King’s College London, London, UK.
  • 2020.09 - 2023.06, Master, with 4.00/4.00 GPA (ranking 1/155), Tsinghua University, Beijing, China.
  • 2016.09 - 2020.06, Bachelor (w. Honor), with 3.83/4.00 GPA (ranking 1/51), Beihang University, Beijing, China.

💻 Internships and RAs

  • 2025.07 - 2025.10, Tuojing Intelligence, Internship: Simulation and Generation in Autonomous Driving.
  • 2023.07 - 2025.07, Tsinghua University, Research Assistant: Perception, Planning and Simulation in Autonomous Driving.
  • 2022.06 - 2022.10, SenseTime Research, Internship: Perception in Unmanned Aerial Vehicle.

💬 Services

  • Since 2024, Conference Reviewer of ECCV, ICLR, NIPS, AAAI, CVPR, etc.
  • Since 2023, Journal Reviewer of Advanced Engineering Informatics, Journal of Industrial Information Integration, Journal of Real-Time Image Processing, The Visual Computer, Signal, Image and Video Processing, etc.
  • 2025.06, Area Chair, CVPR 2025 Workshop on Multi-Agent Embodied Intelligent System.
  • 2024.10, Program Committee Member, ECCV 2024 Workshop on Coop-Intelligence.