Optical Adversarial

Optical adversarial attacks exploit the vulnerabilities of computer vision systems by subtly manipulating the light reaching a camera sensor, causing misclassification or inaccurate depth estimation. Current research focuses on developing increasingly stealthy attacks using techniques like modulated LEDs, strategically placed lenses, and even naturally occurring phenomena like shadows, targeting applications such as autonomous driving and facial recognition. These attacks highlight critical security risks in deploying deep learning models in real-world scenarios, prompting investigations into robust countermeasures at both the hardware and software levels. The ultimate goal is to improve the resilience of computer vision systems against these sophisticated, often imperceptible, manipulations.

Papers

October 31, 2024

Optical Lens Attack on Monocular Depth Estimation for Autonomous Driving
Ce Zhou (1), Qiben Yan (1), Daniel Kent (1), Guangjing Wang (2), Weikang Ding (1), Ziqi Zhang (3), Hayder Radha (1) ((1) Michigan State University, (2) University of South Florida, (3) Peking University)
Autonomous Driving Autonomous Vehicle Monocular Depth Estimation Camera Lens Optical Adversarial

September 25, 2024

Optical Lens Attack on Deep Learning Based Monocular Depth Estimation
Ce Zhou (1), Qiben Yan (1), Daniel Kent (1), Guangjing Wang (1), Ziqi Zhang (2), Hayder Radha (1) ((1) Michigan State University, (2) Peking University)
Depth Estimation Monocular Depth Estimation Camera Lens Vision Based Optical Adversarial

July 10, 2024

Invisible Optical Adversarial Stripes on Traffic Sign against Autonomous Vehicles
Dongfang Guo, Yuting Wu, Yimin Dai, Pengfei Zhou, Xin Lou, Rui Tan
Computer Vision Autonomous Vehicle Image Sensor Rolling Shutter Traffic Sign Recognition Traffic Sign Optical Adversarial Adversarial Disturbance

October 22, 2023

Imperceptible CMOS camera dazzle for adversarial attacks on deep neural networks
Zvi Stein, Adrian Stern
Deep Neural Network Adversarial Attack Invisible Attack Optical Adversarial

July 14, 2023

RFLA: A Stealthy Reflected Light Adversarial Attack in the Physical World
Donghua Wang, Wen Yao, Tingsong Jiang, Chao Li, Xiaoqian Chen
Adversarial Attack Physical World Adversarial Pattern Optical Adversarial Based Physical Attack

March 22, 2023

State-of-the-art optical-based physical adversarial attacks for deep learning computer vision systems
Junbin Fang, You Jiang, Canjian Jiang, Zoe L. Jiang, Siu-Ming Yiu, Chuanyi Liu
Adversarial Attack Vision System Optical Adversarial

November 23, 2022

Privacy-Enhancing Optical Embeddings for Lensless Classification
Eric Bezzam, Martin Vetterli, Matthieu Simeoni
Visual Privacy Lensless Imaging Dynamic Mask Optical Adversarial

May 26, 2022

Physical-World Optical Adversarial Attacks on 3D Face Recognition
Yanjie Li, Yiquan Li, Xuelong Dai, Songtao Guo, Bin Xiao
Adversarial Attack 3D Face 3D Adversarial Optical Adversarial 3D Face Recognition

March 8, 2022

Shadows can be Dangerous: Stealthy and Effective Physical-world Adversarial Attack by Natural Phenomenon
Yiqi Zhong, Xianming Liu, Deming Zhai, Junjun Jiang, Xiangyang Ji
Adversarial Attack Adversarial Example Rare Event Quantum Shadow Physical World Attack Optical Adversarial