# Exploring the Vulnerability of Single Shot Module in Object Detectors via Imperceptible Background Patches

arXiv: https://arxiv.org/abs/1809.05966 · HTML: https://mlsanity.com/paper/1809.05966/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 16 Sep 2018.

## Built on by (most cited first, 12 of 12)

- [Segment and Complete: Defending Object Detectors against Adversarial Patch Attacks with Robust Patch Detection](https://mlsanity.com/paper/2112.04532/) (2021)
- [Robust and Accurate Object Detection via Adversarial Learning](https://mlsanity.com/paper/2103.13886/) (2021)
- [The SARAS Endoscopic Surgeon Action Detection (ESAD) dataset: Challenges and methods](https://mlsanity.com/paper/2104.03178/) (2021)
- [Adversarial Feature Augmentation and Normalization for Visual Recognition](https://mlsanity.com/paper/2103.12171/) (2021)
- [DomainForensics: Exposing Face Forgery across Domains via Bi-directional Adaptation](https://mlsanity.com/paper/2312.10680/) (2023)
- [A Survey and Evaluation of Adversarial Attacks for Object Detection](https://mlsanity.com/paper/2408.01934/) (2024)
- [Simple Transparent Adversarial Examples](https://mlsanity.com/paper/2105.09685/) (2021)
- [Embodied Active Defense: Leveraging Recurrent Feedback to Counter Adversarial Patches](https://mlsanity.com/paper/2404.00540/) (2024)
- [EVD4UAV: An Altitude-Sensitive Benchmark to Evade Vehicle Detection in UAV](https://mlsanity.com/paper/2403.05422/) (2024)
- [SSMI: How to Make Objects of Interest Disappear without Accessing Object Detectors?](https://mlsanity.com/paper/2206.10809/) (2022)
- [Physical world assistive signals for deep neural network classifiers -- neither defense nor attack](https://mlsanity.com/paper/2105.00622/) (2021)
- [Towards Improving Robustness Against Common Corruptions in Object Detectors Using Adversarial Contrastive Learning](https://mlsanity.com/paper/2311.07928/) (2023)

## Abstract (arXiv)

Recent works succeeded to generate adversarial perturbations on the entire image or the object of interests to corrupt CNN based object detectors. In this paper, we focus on exploring the vulnerability of the Single Shot Module (SSM) commonly used in recent object detectors, by adding small perturbations to patches in the background outside the object. The SSM is referred to the Region Proposal Network used in a two-stage object detector or the single-stage object detector itself. The SSM is typically a fully convolutional neural network which generates output in a single forward pass. Due to the excessive convolutions used in SSM, the actual receptive field is larger than the object itself. As such, we propose a novel method to corrupt object detectors by generating imperceptible patches only in the background. Our method can find a few background patches for perturbation, which can effectively decrease true positives and dramatically increase false positives. Efficacy is demonstrated on 5 two-stage object detectors and 8 single-stage object detectors on the MS COCO 2014 dataset. Results indicate that perturbations with small distortions outside the bounding box of object region can still severely damage the detection performance.