Zhiqiang Liu

University of Salerno, University of Melbourne

Papers

Total Citations

H-Index

About

Zhiqiang Liu is a pioneering researcher at the intersection of computer vision and robotics, with a primary focus on embodied perception and active camera systems. His most impactful work, "Active Camera Relocalization from a Single Reference Image without Hand-Eye Calibration" (2018, 27 citations), tackles the challenging problem of determining a camera's 6D pose relative to a single reference image—a task critical for autonomous navigation and robotic manipulation. Liu's key contribution lies in eliminating the need for expensive, error-prone hand-eye calibration, proposing a method that enables a robotic arm to actively reposition a camera to match a reference viewpoint with high precision. This work addresses a fundamental gap in active vision, where cameras must not only see but also physically move to understand their environment. Earlier, in his 2002 paper "Embodied computer vision for mobile robots" (4 citations), Liu explored the philosophical underpinnings of purposive vision, arguing that perception is inherently tied to an agent's physical embodiment and actions. His research bridges low-level vision techniques with high-level robotic control, offering practical solutions for real-world deployment in unstructured environments.