Hacking Object Detectors Is Just Like Training Neural Networks

Explore a novel perspective on object detector deception techniques in this 31-minute conference talk from the Hack In The Box Security Conference. Learn how the process of attacking object detectors shares intrinsic similarities with neural network training. Discover how applying techniques from neural network training can improve the performance of adversarial attacks on state-of-the-art object detection systems. Gain insights from AI security experts Jay Xiong, Dr. Tao Wei, Hao Xin, Wang Yang, and Dou Goodman from Baidu X-Lab as they delve into deep learning security, privacy, and IoT. Understand the implications for AI safety and the potential vulnerabilities in object detection systems used in various applications.

#HITBGSEC COMMSEC: Hacking Object Detectors Is Just Like Training Neural Networks - Jay Xiong