Yumin Sun
Papers
1
Total Citations
11
H-Index
1
About
Yumin Sun is a researcher in 3D computer vision and geometric modeling, with a focus on primitive detection and scene understanding from unorganized point clouds. Their most influential work, "From Planes to Corners: Multi-Purpose Primitive Detection in Unorganized 3D Point Clouds" (2020, 11 citations), introduces a segmentation-free method for jointly estimating orthogonal planes, their intersection lines, relationship graphs, and corners formed by three orthogonal planes. This unified approach under orthogonality constraints enables a wide range of applications, including semantic scene parsing, architectural reconstruction, and robotics perception. Sun's contribution lies in bridging low-level geometric primitive extraction with high-level structural reasoning, offering a robust framework that operates directly on raw point clouds without prior segmentation. This work has been cited for its practical utility in indoor modeling and autonomous navigation. Sun's research continues to advance the integration of geometric priors into deep learning pipelines, making their work valuable for students and researchers exploring efficient 3D scene understanding.
Research Focus
Key Achievements
Top Papers
- 1