Efficient Adversarial Attacks for Visual Object Tracking

Liang, Siyuan; Wei, Xingxing; Yao, Siyuan; Cao, Xiaochun

doi:10.1007/978-3-030-58574-7_3

Cited by 50 publications

(15 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The loss is designed to cool down the hot target regions and forcing the bounding boxes to shrink during online tracking. A Fast-Attack-Network is also developed in [208] to attack the trackers based on Siamese network. In [209], the authors introduced an adversarial pattern that can be printed on a poster in the physical world.…”

Section: A Object Detection and Trackingmentioning

confidence: 99%

Threat of Adversarial Attacks on Deep Learning in Computer Vision: A Survey

2018

View full text Add to dashboard Cite

Deep learning is at the heart of the current rise of artificial intelligence. In the field of Computer Vision, it has become the workhorse for applications ranging from self-driving cars to surveillance and security. Whereas deep neural networks have demonstrated phenomenal success (often beyond human capabilities) in solving complex problems, recent studies show that they are vulnerable to adversarial attacks in the form of subtle perturbations to inputs that lead a model to predict incorrect outputs. For images, such perturbations are often too small to be perceptible, yet they completely fool the deep learning models. Adversarial attacks pose a serious threat to the success of deep learning in practice. This fact has recently lead to a large influx of contributions in this direction. This article presents the first comprehensive survey on adversarial attacks on deep learning in Computer Vision. We review the works that design adversarial attacks, analyze the existence of such attacks and propose defenses against them. To emphasize that adversarial attacks are possible in practical conditions, we separately review the contributions that evaluate adversarial attacks in the real-world scenarios. Finally, drawing on the reviewed literature, we provide a broader outlook of this research direction.

show abstract

Section: A Object Detection and Trackingmentioning

confidence: 99%

Threat of Adversarial Attacks on Deep Learning in Computer Vision: A Survey

2018

View full text Add to dashboard Cite

show abstract

“…We compare the attack method proposed in this paper with the latest attack methods, including the AlexNet-based untargeted attack method CSA [55], and two other targeted attack methods: the AlexNet-based FAN [6] and the ResNet50based TTP [5] (Figure 10). We report the accuracy scores with respect to the real/fake trajectories in table VII.…”

Section: F Comparison With Other Methodsmentioning

confidence: 99%

“…Due to the limitations of the algorithm design, CSA cannot perform targeted attacks for Siamese trackers. Recently, Liang et al proposed a fast attack network (FAN) [6] for attacking SiamFC trackers. To perform untargeted attacks, FAN proposes a drift loss that shifts the tracker's prediction of the target's position.…”

Section: F Comparison With Other Methodsmentioning

confidence: 99%

“…Recently, the robustness of Siamese trackers has attracted much attention in the sense of testing their vulnerability to adversarial attacks, and the focus has been directed towards more efficient and low-cost attacks [5], [6], [7], [8]. However, these attack methods still need to craft the perturbations for each video independently based on either iterative optimization or adversarial network inference, for which the adequacy of additional computational resources may be not ensured by the computationally intensive tracking systems.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Simple and Strong Baseline for Universal Targeted Attacks on Siamese Visual Tracking

Shi

Gao

et al. 2022

IEEE Trans. Circuits Syst. Video Technol.

View full text Add to dashboard Cite

Siamese trackers are shown to be vulnerable to adversarial attacks recently. However, the existing attack methods craft the perturbations for each video independently, which comes at a non-negligible computational cost. In this paper, we show the existence of universal perturbations that can enable the targeted attack, e.g., forcing a tracker to follow the groundtruth trajectory with specified offsets, to be video-agnostic and free from inference in a network. Specifically, we attack a tracker by adding a universal imperceptible perturbation to the template image and adding a fake target, i.e., a small universal adversarial patch, into the search images adhering to the predefined trajectory, so that the tracker outputs the location and size of the fake target instead of the real target. Our approach allows perturbing a novel video to come at no additional cost except the mere addition operations -and not require gradient optimization or network inference. Experimental results on several datasets demonstrate that our approach can effectively fool the Siamese trackers in a targeted attack manner. We show that the proposed perturbations are not only universal across videos, but also generalize well across different trackers. Such perturbations are therefore doubly universal, both with respect to the data and the network architectures. We will make our code publicly available.

show abstract

“…vestigate the robustness of visual tracking algorithms with deep models, recent approaches [3,44,16,24] assume that the model structures of deep tracking algorithms are known and carry out white-box attack on them. Despite the demonstrated promising results, the concrete structures and parameters of deep trackers are barely known in real applications.…”

Section: Siamrpn++mentioning

confidence: 99%

IoU Attack: Towards Temporally Coherent Black-Box Adversarial Attack for Visual Object Tracking

Jia¹,

Song²,

Ma³

et al. 2021

Preprint

View full text Add to dashboard Cite

Adversarial attack arises due to the vulnerability of deep neural networks to perceive input samples injected with imperceptible perturbations. Recently, adversarial attack has been applied to visual object tracking to evaluate the robustness of deep trackers. Assuming that the model structures of deep trackers are known, a variety of white-box attack approaches to visual tracking have demonstrated promising results. However, the model knowledge about deep trackers is usually unavailable in real applications. In this paper, we propose a decision-based black-box attack method for visual object tracking. In contrast to existing black-box adversarial attack methods that deal with static images for image classification, we propose IoU attack that sequentially generates perturbations based on the predicted IoU scores from both current and historical frames. By decreasing the IoU scores, the proposed attack method degrades the accuracy of temporal coherent bounding boxes (i.e., object motions) accordingly. In addition, we transfer the learned perturbations to the next few frames to initialize temporal motion attack. We validate the proposed IoU attack on stateof-the-art deep trackers (i.e., detection based, correlation filter based, and long-term trackers). Extensive experiments on the benchmark datasets indicate the effectiveness of the proposed IoU attack method. The source code is available at https://github.com/VISION-SJTU/IoUattack.

show abstract

Efficient Adversarial Attacks for Visual Object Tracking

Cited by 50 publications

References 28 publications

Threat of Adversarial Attacks on Deep Learning in Computer Vision: A Survey

Threat of Adversarial Attacks on Deep Learning in Computer Vision: A Survey

A Simple and Strong Baseline for Universal Targeted Attacks on Siamese Visual Tracking

IoU Attack: Towards Temporally Coherent Black-Box Adversarial Attack for Visual Object Tracking

Contact Info

Product

Resources

About