Zhangjin Huang

University of Science and Technology of China

Papers

Total Citations

H-Index

About

Zhangjin Huang is a researcher whose work is centered on advancing 3D computer vision, with a particular focus on 6D object pose estimation—a critical capability for robotic manipulation and augmented reality. Huang’s major contributions lie in developing novel deep learning architectures that leverage cross-modal and geometric reasoning. In the 2021 paper “CMA: Cross-modal attention for 6D object pose estimation” (22 citations), Huang introduced a cross-modal attention mechanism to effectively fuse RGB and depth information, significantly improving pose estimation accuracy. Building on this, the 2023 work “Learning geometric consistency and discrepancy for category-level 6D object pose estimation from point clouds” (22 citations) tackles the challenging task of estimating poses for unseen object instances by exploiting geometric features from point clouds, moving beyond reliance on visual RGB cues. This work addresses a fundamental problem in robotics, where depth data often receives less attention despite its value. Huang’s research is impactful, with each of these key papers accumulating 22 citations, demonstrating growing recognition in the field. By pioneering methods that combine attention-based learning with geometric consistency, Huang is helping to make 6D pose estimation more robust and applicable to real-world robotic systems.