Ying Shan

Papers

Total Citations

H-Index

About

Ying Shan is a leading researcher in computer vision and autonomous driving, with a focus on 3D object detection from LiDAR point clouds. Their most notable contribution is the pioneering work "SCP: Scene Completion Pre-training for 3D Object Detection" (2023), which addresses a critical bottleneck in the field: the heavy reliance on expensive and error-prone 3D bounding box annotations. By introducing a scene completion pre-training paradigm, Shan’s method enables 3D detectors to learn robust representations from unlabeled data, significantly reducing the need for manual labeling while improving detection accuracy. This work has garnered early citations, reflecting its growing influence in the autonomous driving community. Shan’s research sits at the intersection of self-supervised learning, 3D scene understanding, and robotics, with the potential to accelerate the development of safer, more scalable perception systems. Their contributions are particularly impactful for students and researchers seeking to reduce annotation costs in real-world deployment scenarios, marking Shan as a rising innovator in 3D vision and autonomous systems.