Zhiheng Fu

The University of Western Australia, National University of Defense Technology

Papers

Total Citations

H-Index

About

Zhiheng Fu is a researcher whose work sits at the critical intersection of 3D perception, sensor fusion, and robotic localization. His primary research areas include depth estimation, multi-modal sensor fusion, and large-scale visual SLAM. Fu’s most significant contribution is the development of SLFNet, a pioneering Stereo and LiDAR Fusion Network for depth completion. This work, which has garnered 15 citations, directly addresses the real-time demands of autonomous driving and robotic perception by intelligently combining the complementary strengths of stereo cameras and LiDAR sensors to produce dense, precise depth maps. Building on this, his work on multi-stage information diffusion for joint depth and surface normal estimation (9 citations) further advances scene understanding. Earlier in his career, Fu also tackled the challenge of visual loop closure detection in large-scale environments with his work on simultaneous context feature learning and hashing. By addressing the discriminability of descriptors in repetitive settings, this research has implications for robust pose tracking and relocalization in augmented reality and robotics. Through these contributions, Zhiheng Fu is helping to build the foundational perception systems that will enable safer autonomous navigation and more immersive AR experiences.