Fan Tang

Chinese Academy of Sciences, Nanyang Technological University, Fordham University

Papers

Total Citations

H-Index

About

Fan Tang’s research lies at the intersection of 3D computer vision and autonomous robotics, with a focus on enabling machines to perceive and navigate complex environments with greater precision and reliability. A key contribution is the development of the **Semantic-Context Graph Network** for point-based 3D object detection, which tackles the persistent challenge of semantic ambiguity—such as shape symmetries and occlusion—in indoor scenes. This work, published in 2023, has already garnered 16 citations, reflecting its timely impact on applications in augmented reality, autonomous driving, and robotics. Tang’s earlier foundational work on **pose-invariant feature extraction** using a modified scale-space approach (2004, 15 citations) provided a robust method for extracting landmarks from 2D laser range data, a critical component for feature-based SLAM in mobile robotics. Further, Tang has advanced the field by introducing **formal performance guarantees** for behavior-based localization missions (2016), moving beyond empirical validation to provide provable assurances for robot navigation tasks. This work, integrated with the MissionLab/VIPARS framework, demonstrates a rare combination of theoretical rigor and practical deployment. Tang’s career thus bridges classical robotics perception with modern deep learning, offering students a model of sustained, impactful research.