Pan Zhou
Papers
1
Total Citations
2
H-Index
1
About
Pan Zhou is a researcher whose work spans adversarial machine learning, 3D computer vision, and AI security. His research has focused particularly on the vulnerabilities of deep learning models operating on 3D point cloud data — a critical concern as depth sensors become increasingly integrated into safety-critical systems such as autonomous driving and robot navigation. His notable work, "3DHacker: Spectrum-based Decision Boundary Generation for Hard-label 3D Point Cloud Attack" (2023), represents a meaningful contribution to understanding how 3D point cloud models can be compromised under challenging hard-label black-box conditions, addressing a gap left by prior work that relied on more permissive white-box settings. By leveraging spectral analysis to generate adversarial decision boundaries, Zhou's approach advances the field's understanding of real-world attack scenarios where an adversary has only limited access to a model's outputs. Though early in citation accumulation with 2 citations, this work addresses timely and consequential security questions as 3D perception systems proliferate across industries. Zhou's research contributes to the broader effort of building more robust and trustworthy AI systems for real-world deployment.
Research Focus
Key Achievements
Top Papers
- 1