Hongmin Liu

University of Science and Technology Beijing, Ministry of Education of the People's Republic of China

Papers

Total Citations

H-Index

About

Hongmin Liu is a leading researcher in computer vision and robotics, with a focus on 3D scene reconstruction, visual localization, and autonomous driving perception. Their work addresses critical challenges in enabling machines to perceive and navigate complex environments with high precision. A standout contribution is the development of a complete and accurate indoor scene capturing and reconstruction system using a drone and a robot (17 citations), which set a new benchmark for coverage and fidelity in image-based 3D modeling. In autonomous driving, Liu introduced a bidirectional agent-map interaction learning framework for trajectory prediction (11 citations), advancing the modeling of dynamic agent-environment interactions. Their innovations in visual localization include a lightweight structured line map method (6 citations) and a task-aligned local feature learning approach (2 citations), both designed to improve robustness and efficiency for applications in robotics and augmented reality. By integrating map-related tasks into feature learning, Liu has pushed the boundaries of reliable camera pose estimation. With a growing citation impact and a focus on practical, deployable systems, Hongmin Liu’s work is shaping the future of intelligent spatial perception.