Hi there! I am Shengpeng Wang, a PhD student at Huazhong University of Science and Technology. I am currently a member of metaiot under the supervision of Prof. Wei Wang. Before that, I obtained my bachelor’s degree at the School of Information and Engineering, Wuhan University of Technology.
My research interest includes World Action Model, SLAM, Gaussian Splatting, 3D Perception & Understanding in adverse scenes, if you have an excellent project for collaboration, please email me!
🔥 News
- 2026.06: 🎉🎉 Our paper about Radar-Inertial calibration and imaging is accepted to ECCV’26.
- 2026.02: 🎉🎉 Our Spatial Uncertainty paper is accepted to CVPR’26 HighLight.
- 2025.11: 🎉🎉 Our 3D Spatial Perception paper is accepted to AAAI’25 oral.
- 2025.06: 🎉🎉 Our radar-based SLAM paper is accepted to ICCV’25 poster.
- 2025.04: 🎉🎉 Our 3D Perception paper is accepted to IJCAI’25 oral.
📝 Publications
Mitigating Calibration Plights via Arbitrary Rotational Manipulation
Chunshen Li, Shengpeng Wang, Zitao Ye, Wei Wang
- This work studies practical calibration challenges in radar-inertial systems and investigates a flexible rotational manipulation strategy for improving calibration and imaging reliability.
RaUF: Learning the Spatial Uncertainty Field of Radar
Shengpeng Wang, Kuangyu Wang, Wei Wang
- We introduce a novel radar perception framework that learns anisotropic geometric uncertainty calibration to resolve the inherently ill-posed geometric inference of low-fidelity radar space.
S3E: Self-Supervised State Estimator for Radar-Inertial System
Shengpeng Wang, Yulong Xie, Qing Liao, Yue Cao, Wei Wang
- The pioneering endeavor to achieve state estimation and landmark extraction from radar signal spectra and inertial data in a self-supervised manner.
SDDiff: Boost Radar Perception via Spatial-Doppler Diffusion
Shengpeng Wang, Xin Luo, Yulong Xie, Wei Wang
- This is the first attempt to simultaneously attain dense point clouds and accurate ego velocity from the Spatial-Doppler Domain Representation
RaLD: Generating High-Resolution 3D Radar Point Clouds with Latent Diffusion
Ruijie Zhang, Bixin Zeng, Shengpeng Wang, Fuhui Zhou, Wei Wang
- This introduces a new paradigm for spectrum-conditioned 3D point cloud generation from radar to enhance both the fidelity and structural integrity of point clouds.
🌐 Academic Service
- Conference and Journal Reviewer: CVPR, ICCV, ICML, NeurIPS, IJCAI, AAAI, ICRA, RA-L, TMC.
- Review Areas: embodied spatial intelligence, 3D perception, robotic perception, multimodal fusion, radar perception, autonomous localization, and spatial reconstruction.
- Conference Volunteer: International Joint Conference on Artificial Intelligence, 2025.
- Invited Academic Activities: Ali Star Academic Seminar on Foundation and Multimodal Models; Invited research presentation at ZANQI Technology; Invited research presentation at ArtInHCI 2026.
🎖 Honors and Awards
- 2025.10 Huawei Scholarship Awarded in HUST
- 2022-2025 Ph.D. Scholarships.
- 2022.10 Second Prize of Huawei Cup Chinese Graduate Mathematical Modeling Competition
- 2022.04 First Prize of National College Student Mathematics Competition
- 2021.04 Finalist of Mathematical Contest In Modeling
- 2020.09 First Prize of the National College Student Mathematical Modeling Competition of High Education Society Cup.
📖 Educations
- 2022.09 - (now), Huazhong University of Science and Technology, Ph.D. Student, Information and Communication Engineering.
- 2018.09 - 2022.06, Wuhan University of Technology, B.E., Communication Engineering.