Haobo Jiang

Nanjing University of Science and Technology

Papers

Total Citations

H-Index

About

Haobo Jiang is a rising researcher in computer vision and robotics, with a primary focus on point cloud registration (PCR)—a foundational challenge for 3D perception and autonomous systems. His most notable contribution, the Salient Geometric Network (SGNet), addresses a core difficulty in PCR: identifying semantically and geometrically consistent keypoints across disparate 3D scans. By designing a network that explicitly learns salient geometric features, Jiang’s work improves the robustness and accuracy of aligning point clouds, which is critical for applications like SLAM, 3D reconstruction, and object recognition. Although his most-cited paper is recent (2024), its 2 citations reflect early-stage recognition in a fast-moving field. Jiang’s research stands out for tackling the “saliency” problem—moving beyond brute-force matching to intelligent, structure-aware correspondence. His work is particularly relevant for students and engineers working on LiDAR-based perception or 3D scene understanding, offering a principled approach to making PCR more reliable in noisy, partial, or occluded environments. As the field pushes toward real-time, high-fidelity 3D mapping, Jiang’s geometric deep learning methods are poised to become building blocks for next-generation spatial AI systems.