Zhaoxuan Zhang

Papers

Total Citations

H-Index

About

Zhaoxuan Zhang is a researcher in 3D computer vision, with a primary focus on point cloud processing and semantic scene understanding. Their most cited work, "Point cloud semantic scene segmentation based on coordinate convolution" (2020, 12 citations), addresses a fundamental challenge in the field: applying convolution operations to irregular and unordered point cloud data. This paper introduces a novel coordinate convolution approach that enables more effective semantic segmentation of 3D scenes, a critical capability for applications in autonomous robotics, augmented reality, and 3D mapping. By tackling the inherent difficulties of point cloud convolution, Zhang's work contributes to bridging the gap between traditional 2D image processing and the complex, unstructured nature of 3D data. Their research is particularly relevant for advancing scene-level understanding, where accurate segmentation of objects and surfaces in point clouds is essential. Though early in their career, Zhang's contributions demonstrate a clear commitment to solving foundational problems in 3D computer vision, laying groundwork for more robust and efficient spatial perception systems.