Profile
Wanxiang Qin is currently a full-time faculty member at the School of Arts and Design, Yulin Normal University.
Education
- M.E., Architecture, Inner Mongolia University of Technology, 2020—2023.
- B.E., Architecture, Guangxi University of Science and Technology, 2015—2020.
Research Fields
computer vision , 3D gaussian splatting, Unreal Engine rendering pipeline and related fields.
Publications
- Qin Wanxiang,Hoi Leong Lee. Material-Aware 3D Gaussian Splatting: Learning Spatially Varying BRDF with Physics-Based Material Priors(CSCWD2025)(CCF C)
- WanXiang Qin, Ji Li, Peng Qu, Zhen Tian, Liang Tan (2025). PPA-Enhanced YOLOv10: Real-Time Detection of Small UAVs in Complex Environments. 2025 2nd International Conference on Autonomous Driving and Intelligent Sensing Technology (ADIST 2025).
- MingChen, WanXiang Qin, Tomato Ripeness Detection Method Based on FasterNet Block and Attention Mechanism (AIP Advances) (SCI Q4).
- Qin Wanxiang (2023). Research on Multi Objective Optimization of High rise Residential Buildings in Severe Cold Regions Based on Energy Consumption and Solar Thermal Performance, Master's Thesis, Inner Mongolia University of Technology.
Work Experience
- Shenzhen Xkool Technology, 2021.06–2021.12. — Code and architecture construction of Python city generation; built a visual system for the digital architecture department using Python.
- Mitu Digital Technology, 2022.01–2022.07. — Wrote interactive functions and character controllers using UE Blueprints and Unity C#; authored shaders with UE material nodes and Unity Shader Graph.
- Shenzhen Qijing Forest Technology, 2022.07–2022.12. — Implemented data interaction between front-end JavaScript and UE in UE cloud rendering platform; automated import of user uploaded models in UE cloud rendering platform.
- Yulin Normal University School, 2023.06–PRESENT. — Full-time teacher.
3DGS Display(ThreeJS · Gaussian Splatting)
Instructions
This display is built upon the ThreeJS and GaussianSplats3D libraries, directly reading `.ply/.splat/.ksplat` files for Gaussian point rendering, and offering interactive viewing. If manual data construction is required, refer to the JSON example (optional):
{
"gaussians": [
{ "position": [x, y, z], "color": [r, g, b], "scale": s, "sigma": [sx, sy], "rot": r },
{ "position": [x, y, z], "color": [r, g, b], "scale": s }
]
}The color range is 0–1, and the scale is the basic size of the particles; optional `sigma=[sx,sy]` and `rot` (radians) are used for anisotropic ellipses and rotations. In production environments, it is recommended to use `.ksplat` for better loading and rendering performance.
Contact
- EMAIL:msxy20230097@ylu.edu.cn
- WEBSITE:wanxiangqin.github.io