Towards Robust Object Detection: Identifying and Removing Backdoors via Module Inconsistency Analysis

TL;DR

This paper introduces a novel framework for detecting and removing backdoors in object detection models by analyzing inconsistencies between local modules, significantly improving security against backdoor attacks.

Contribution

It presents the first method tailored for two-stage object detectors that detects backdoors through module inconsistency analysis and effectively removes them with minimal impact on accuracy.

Findings

Achieves 90% backdoor removal rate improvement over baselines

Limits clean data accuracy loss to less than 4%

First approach addressing backdoor detection and removal in two-stage detectors

Abstract

Object detection models, widely used in security-critical applications, are vulnerable to backdoor attacks that cause targeted misclassifications when triggered by specific patterns. Existing backdoor defense techniques, primarily designed for simpler models like image classifiers, often fail to effectively detect and remove backdoors in object detectors. We propose a backdoor defense framework tailored to object detection models, based on the observation that backdoor attacks cause significant inconsistencies between local modules' behaviors, such as the Region Proposal Network (RPN) and classification head. By quantifying and analyzing these inconsistencies, we develop an algorithm to detect backdoors. We find that the inconsistent module is usually the main source of backdoor behavior, leading to a removal method that localizes the affected module, resets its parameters, and fine-tunes the model on a small clean dataset. Extensive experiments with state-of-the-art two-stage object detectors show our method achieves a 90% improvement in backdoor removal rate over fine-tuning baselines, while limiting clean data accuracy loss to less than 4%. To the best of our knowledge, this work presents the first approach that addresses both the detection and removal of backdoors in two-stage object detection models, advancing the field of securing these complex systems against backdoor attacks.