Weien Zhou

Papers

Total Citations

H-Index

About

Weien Zhou is a researcher whose work sits at the intersection of 3D computer vision and adversarial machine learning, with a particular focus on the security of point cloud perception systems. His most cited work, "Gradient-based sparse voxel attacks on point cloud object detection" (2024, 11 citations), introduces a novel adversarial attack that exploits the voxel-based representations commonly used in LiDAR-based object detectors. By crafting sparse, gradient-driven perturbations, Zhou demonstrates how even minimal modifications to point cloud data can fool state-of-the-art detection models—a critical vulnerability for autonomous driving and robotics. This contribution has quickly garnered attention, highlighting the growing importance of robustness in 3D deep learning. Zhou’s research is notable for its practical implications: his attack methods expose real-world risks while also providing a foundation for developing more resilient detection architectures. As a rising voice in adversarial 3D vision, his work bridges the gap between theoretical security analysis and applied perception systems, making him a key figure to watch in the ongoing effort to safeguard AI-driven spatial understanding.