Explore a 16-minute IEEE conference talk on certifiably robust object detection techniques that counter patch hiding attacks. Delve into the innovative ObjectSeeker method, which employs patch-agnostic masking to enhance detection reliability. Learn from Princeton University researchers Chong Xiang, Alexander Valtchanov, Saeed Mahloujifar, and Prateek Mittal as they present their findings on improving object detection systems' resilience against adversarial manipulations.
ObjectSeeker - Certifiably Robust Object Detection against Patch Hiding Attacks via Patch-agnostic Masking
Overview
Syllabus
Certifiably Robust Object Detection against Patch Hiding Attacks via Patch-agnostic Masking
Taught by
IEEE Symposium on Security and Privacy
Tags
Related Courses
-
A Deep Learning Approach to Fast, Format-Agnostic Detection of Malicious Web Content
-
GraphSPD - Graph-Based Security Patch Detection with Enriched Code Semantics
-
Goshawk - Hunting Memory Corruptions via Structure-Aware and Object-Centric Memory Operation Synopsis
-
RAB - Provable Robustness Against Backdoor Attacks
-
Distillation as a Defense to Adversarial Perturbations against Deep Neural Networks
-
Model Stealing Attacks Against Inductive Graph Neural Networks
