Haoran Pan

Nanjing University of Aeronautics and Astronautics

Papers

Total Citations

H-Index

About

Haoran Pan is a rising researcher in computer vision and robotics, with a focused interest in 3D geometric deep learning and 6D object pose estimation. His most cited work, “SO(3)‐Pose: SO(3)‐Equivariance Learning for 6D Object Pose Estimation” (2022), introduces a novel framework that leverages SO(3) equivariance to robustly fuse RGB appearance and depth geometry for precise object pose estimation. This contribution addresses a critical challenge in robotic manipulation—how to align visual and spatial information under rotational transformations—by embedding group-equivariant learning directly into the pose estimation pipeline. Though early in his career, Pan’s work has already garnered attention (4 citations), signaling its relevance to the growing field of equivariant neural networks. His research stands at the intersection of theoretical symmetry and practical robotics, offering a principled approach to improving grasping and interaction in unstructured environments. As the demand for reliable object manipulation in autonomous systems increases, Pan’s contributions to equivariant pose learning position him as a promising voice in the next generation of computer vision researchers.