A survey on adversarial attacks in computer vision: Taxonomy, visualization and future directions

Long, Teng; Gao, Qi; Xu, Lili; Zhou, Zhangbing

doi:10.1016/j.cose.2022.102847

Cited by 38 publications

(6 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Sadeghi et al [34] summarizes the basic concepts, analyzes datasets and ML architectures, providing awareness of the adversary strategies and the defense responses to them, and proposes a comprehensive system-drive taxonomy for the latter. Long et al [22] discusses a set of preliminary concepts of Computer Vision and adversarial context, providing a set of adversarial attacks grouped by adversary goals and capabilities, and propose a set of research directions that readers can use to continue the development of robust networks. Liang et al [23] discuss the most significant attacks and defenses in the literature, with the latter being grouped by the underlying technique, finishing with the presentation of existing challenges in the adversarial context.…”

Section: Background For Adversarial Attacksmentioning

confidence: 99%

“…Previous surveys [22]- [24] do not provide a state-of-theart comparison, area-specific metrics, and architectures commonly used, hindering the initiation among new researchers. This survey provides a comprehensive review of the legacy and recent advances, such as Adversarial Purification and Adversarial Attacks in Vision Transformers (ViTs), presents datasets and metrics in this context, provides results for multiple datasets, using the de facto standard attack (Auto-Attack [25]), and summarizes approaches that have official or unofficial implementations (duly identified) to stimulate the research of new mechanisms.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

How Deep Learning Sees the World: A Survey on Adversarial Attacks & Defenses

Costa,

Roxo,

Proença

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Deep Learning is currently used to perform multiple tasks, such as object recognition, face recognition, and natural language processing. However, Deep Neural Networks (DNNs) are vulnerable to perturbations that alter the network prediction, named adversarial examples, which raise concerns regarding the usage of DNNs in critical areas, such as Self-driving Vehicles, Malware Detection, and Healthcare. This paper compiles the most recent adversarial attacks in Object Recognition, grouped by the attacker capacity and knowledge, and modern defenses clustered by protection strategies, providing background details to understand the topic of adversarial attacks and defenses. The new advances regarding Vision Transformers are also presented, which have not been previously done in the literature, showing the resemblance and dissimilarity between this architecture and Convolutional Neural Networks. Furthermore, the most used datasets and metrics in adversarial settings are summarized, along with datasets requiring further evaluation, which is another contribution. This survey compares the state-of-the-art results under different attacks for multiple architectures and compiles all the adversarial attacks and defenses with available code, comprising significant contributions to the literature. Finally, practical applications are discussed, and open issues are identified, being a reference for future works.INDEX TERMS Adversarial attacks, adversarial defenses, datasets, evaluation metrics, review, vision transformers.

show abstract

Section: Background For Adversarial Attacksmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

How Deep Learning Sees the World: A Survey on Adversarial Attacks & Defenses

Costa,

Roxo,

Proença

et al. 2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…И в некоторых ситуациях, когда человек легко находит решение, нейронная сеть ошибается, как бы сложно написана она ни была. Более того, большинство нейронных сетей очень легко обмануть [6].…”

Section: Introductionunclassified

Educational and methodological materials of the master class “Adversarial attacks on image recognition neural networks” for students and schoolchildren

Pantiukhin

2023

Informatika i obrazovanie

View full text Add to dashboard Cite

The problem of neural network vulnerability has been the subject of scientific research and experiments for several years. Adversarial attacks are one of the ways to “trick” a neural network, to force it to make incorrect classification decisions. The very possibility of adversarial attack lies in the peculiarities of machine learning of neural networks. The article shows how the properties of neural networks become a source of problems and limitations in their use. The materials of the corresponding researches of the author were used as a basis for the master class “Adversarial attacks on image recognition neural networks”.The article presents the educational materials of the master class: the theoretical background of the class, practical materials (in particular, the attack on a single neuron is described, the fast gradient sign method for attacking a neural network is considered), examples of experiments and calculations (the author uses the convolutional network VGG, Torch and CleverHans libraries), as well as a set of typical errors of students and the teacher’s explanations of how to eliminate these errors. In addition, the result of the experiment is given in the article, and its full code and examples of approbation of the master class materials are available at the above links.The master class is intended for both high school and university students who have learned the basics of neural networks and the Python language, and can also be of practical interest to computer science teachers, to developers of courses on machine learning and artificial intelligence as well as to university teachers.

show abstract

“…Many attack algorithms have been developed so far [2] and applied to various types of data, such as images, speech signals, and texts. Especially for images, we can find various attack algorithms [1,12,8,5,6]. In a recent survey [6], major image attack algorithms are classified into gradient-based, transfer/score-based, decision-based, and approximation-based algorithms.…”

Section: Introductionmentioning

confidence: 99%

Toward Defensive Letter Design

Kataoka,

Kimura,

Uchida

2023

Lecture Notes in Computer Science

View full text Add to dashboard Cite

A major approach for defending against adversarial attacks aims at controlling only image classifiers to be more resilient, and it does not care about visual objects, such as pandas and cars, in images. This means that visual objects themselves cannot take any defensive actions, and they are still vulnerable to adversarial attacks. In contrast, letters are artificial symbols, and we can freely control their appearance unless losing their readability. In other words, we can make the letters more defensive to the attacks. This paper poses three research questions related to the adversarial vulnerability of letter images: (1) How defensive are the letters against adversarial attacks? (2) Can we estimate how defensive a given letter image is before attacks? (3) Can we control the letter images to be more defensive against adversarial attacks? For answering the first and second questions, we measure the defensibility of letters by employing Iterative Fast Gradient Sign Method (I-FGSM) and then build a deep regression model for estimating the defensibility of each letter image. We also propose a two-step method based on a generative adversarial network (GAN) for generating character images with higher defensibility, which solves the third research question.

show abstract

A survey on adversarial attacks in computer vision: Taxonomy, visualization and future directions

Cited by 38 publications

References 35 publications

How Deep Learning Sees the World: A Survey on Adversarial Attacks & Defenses

How Deep Learning Sees the World: A Survey on Adversarial Attacks & Defenses

Educational and methodological materials of the master class “Adversarial attacks on image recognition neural networks” for students and schoolchildren

Toward Defensive Letter Design

Contact Info

Product

Resources

About