Zhihao Cui

Pingdingshan University, Wuhu Hit Robot Technology Research Institute

Papers

Total Citations

H-Index

About

Zhihao Cui is a researcher at the intersection of computer vision, robotics, and intelligent systems, with a primary focus on enabling precise, real-time object manipulation in industrial and sports applications. His work is defined by tackling the practical challenges of robotic perception—from the fast-paced recognition needed in sports to the high-precision grasping required in manufacturing. Cui’s most significant contributions lie in developing novel deep learning frameworks for 6D object pose estimation and instance segmentation, particularly for difficult scenarios involving reflective, textureless, or slender industrial workpieces. His paper "PS6D" introduces a symmetry-aware, point-cloud-based method that directly addresses the failure modes of traditional algorithms in bin-picking tasks. Similarly, "KeypointMask" advances real-time instance segmentation for oblique objects, a critical step for robust robotic grasping. With over 10 citations across his most-cited works, Cui’s research is gaining traction for its practical impact. His earlier work on a "Badminton Recognition and Tracking System" demonstrates a broader interest in applying human-computer interaction to dynamic environments, showcasing his versatility in bridging theoretical advances with real-world robotic systems.