Fangzheng Tian

Shanghai University of Engineering Science

Papers

Total Citations

H-Index

About

Fangzheng Tian is a researcher specializing in 3D computer vision, with a focus on depth estimation, pose estimation, and 3D reconstruction. His work addresses critical challenges in reconstructing accurate 3D scenes from 2D images, particularly in dynamic or texture-poor environments. His most-cited paper, "3D reconstruction with auto-selected keyframes based on depth completion correction and pose fusion" (2021, 8 citations), introduces a novel pipeline that automatically selects optimal keyframes, corrects depth maps using completion techniques, and fuses poses to improve reconstruction robustness. This contribution is significant for applications in robotics, augmented reality, and autonomous navigation, where reliable 3D perception is essential. Tian’s approach stands out for its ability to handle sensor noise and missing data, enhancing the accuracy and efficiency of multi-view reconstruction. His work has been recognized in the computer vision community for its practical impact, and he continues to advance the field by integrating learning-based methods with geometric constraints. With a growing citation record, Tian is establishing himself as a promising researcher in 3D vision, bridging the gap between theoretical algorithms and real-world deployment.