Ziyong Feng

Papers

Total Citations

H-Index

About

Ziyong Feng is a researcher advancing the frontiers of 3D computer vision and robotics, with a primary focus on point cloud registration—a critical task for enabling autonomous systems to perceive and navigate their environments. His most-cited work, "Point Cloud Registration using Representative Overlapping Points" (2021, 23 citations), tackles a persistent challenge in the field: achieving robust registration under partial overlap conditions. While many learning-based methods depend heavily on precise correspondences and falter with incomplete data, Feng’s approach introduces a novel framework that identifies representative overlapping points, significantly improving alignment accuracy and robustness. This contribution addresses a fundamental bottleneck in applications like SLAM, autonomous driving, and 3D reconstruction. Though early in his career, Feng’s work has already garnered attention for its practical relevance, offering a more resilient solution to real-world scenarios where sensor data is often noisy or occluded. His research underscores a commitment to bridging the gap between theoretical advances and deployable robotic perception, marking him as a promising voice in the ongoing evolution of 3D vision technologies.