DetectS ec: Evaluating the robustness of object detection models to adversarial attacks. Issue 9 (8th February 2022)

• Record Type: Journal Article
• Authors: Du, Tianyu; Ji, Shouling; Wang, Bo; He, Sirui; Li, Jinfeng; Li, Bo; Wei, Tao; Jia, Yunhan; Beyah, Raheem; Wang, Ting
• Is Part Of: International journal of intelligent systems. Volume 37:Issue 9(2022)
• Page Start: 6463
• Page End: 6492
• Publication Date: 2022-02-08
• Subjects: adversarial attack -- deep learning -- neural network -- object detection -- robustness evaluation; Artificial intelligence -- Periodicals; Expert systems (Computer science) -- Periodicals; Intelligence artificielle -- Périodiques; Systèmes experts (Informatique) -- Périodiques; 006.3
• DOI: 10.1002/int.22851
• Languages: English
• ISSNs: 0884-8173

Abstract

Abstract: Despite their tremendous success in various machine learning tasks, deep neural networks (DNNs) are inherently vulnerable to adversarial examples, which are maliciously crafted inputs to cause DNNs to misbehave. Intensive research has been conducted on this phenomenon in simple tasks (e.g., image classification). However, little is known about this adversarial vulnerability for object detection, a much more complicated task, which often requires specialized DNNs and multiple additional components. In this paper, we present Detect Sec, a uniform platform for robustness analysis of object detection models. Currently, Detect Sec implements 13 representative adversarial attacks with 7 utility metrics and 13 defenses on 18 standard object detection models. Leveraging Detect Sec, we conduct the first rigorous evaluation of adversarial attacks on the state‐of‐the‐art object detection models. We analyze the impact of the factors including DNN architecture and capacity on the model robustness. We show that many conclusions about adversarial attacks and defenses in image classification tasks do not transfer to object detection tasks, for example, the targeted attack is stronger than the untargeted attack for two‐stage detectors. Our findings will aid future efforts in understanding and defending against adversarial attacks in complicated tasks. In addition, we compare the robustness of different detection models and discuss their relative strengths and weaknesses. The platform Detect Sec will be open source as a unique facility for further research on adversarial attacks and defenses in object detection tasks.