Xusheng Zhu

Papers

Total Citations

H-Index

About

Xusheng Zhu is a researcher in computer vision and 3D perception, with a focus on advancing object pose estimation for robotic and augmented reality applications. His most-cited work, "BOLD3D: A 3D BOLD descriptor for 6DoF pose estimation" (2020), introduces a novel binary descriptor that efficiently encodes local geometry from 3D point clouds, enabling robust and real-time six-degree-of-freedom pose estimation. This contribution addresses critical challenges in handling occlusions and clutter, achieving high accuracy while maintaining computational efficiency—a balance often difficult in practical systems. With 11 citations, the paper has already influenced subsequent work in descriptor-based pose estimation, demonstrating its relevance to the field. Zhu’s research bridges theoretical rigor and applied robotics, offering tools that enhance machine perception in dynamic environments. His work is particularly notable for its potential to improve autonomous manipulation and scene understanding, marking him as a promising voice in 3D computer vision.