Zhixuan Liu

Papers

Total Citations

H-Index

About

Zhixuan Liu is a robotics researcher specializing in 3D vision and robotic manipulation, with a core focus on grasp detection for dexterous, 7-degree-of-freedom (7-DoF) robotic grippers. His work addresses a critical challenge in autonomous robotics: enabling robots to reliably grasp objects in cluttered, real-world environments using incomplete sensor data. Liu’s most influential contribution is the **TransGrasp** framework (2022, 30 citations), which pioneered a multi-scale hierarchical point transformer to capture non-local geometric information from point clouds, significantly improving grasp pose prediction over traditional PointNet++ backbones. He further advanced the field by tackling single-view depth limitations in his 2024 work (12 citations), simulating complete point representations to compensate for occluded geometry. His 2023 study on grasp region exploration (6 citations) introduced strategies for precise, collision-free grasp configuration in dense clutter. Collectively, Liu’s research has pushed the frontier of 7-DoF grasp detection, directly impacting the reliability of robotic manipulation in unstructured settings. His work is essential reading for students and engineers developing next-generation autonomous grasping systems.