Lubing Zhou

Papers

Total Citations

424

H-Index

About

Lubing Zhou is a leading researcher in 3D computer vision and autonomous driving perception, with a focus on LiDAR-based scene understanding. His most influential work, "PointPillars: Fast Encoders for Object Detection From Point Clouds" (2019), has garnered 241 citations and revolutionized real-time 3D object detection by introducing a pillar-based encoding method that balances speed and accuracy, making it a cornerstone for many robotics and autonomous driving pipelines. Zhou further advanced the field with "Panoptic Nuscenes: A Large-Scale Benchmark for LiDAR Panoptic Segmentation and Tracking" (2022, 183 citations), which established a comprehensive benchmark for joint panoptic segmentation and multi-object tracking in urban environments. This work enables robots and automated vehicles to simultaneously understand static scene elements and track dynamic agents with high reliability, leveraging LiDAR’s geometric precision. Zhou’s contributions have provided critical tools for scalable, real-time perception systems, directly impacting autonomous navigation technologies. His research continues to shape how point cloud data is processed for safe and efficient robotic operation in complex, real-world settings.