mlsanityData to 7 Oct 2026

Sparse Adversarial Video Attacks with Spatial Transformations

arXiv 2111.05468 ↗First posted 10 Nov 2021cs.CVcited by 11

Not on the map. Why?

Abstract (arXiv)

In recent years, a significant amount of research efforts concentrated on adversarial attacks on images, while adversarial video attacks have seldom been explored. We propose an adversarial attack strategy on videos, called DeepSAVA. Our model includes both additive perturbation and spatial transformation by a unified optimisation framework, where the structural similarity index (SSIM) measure is adopted to measure the adversarial distance. We design an effective and novel optimisation scheme which alternatively utilizes Bayesian optimisation to identify the most influential frame in a video and Stochastic gradient descent (SGD) based optimisation to produce both additive and spatial-transformed perturbations. Doing so enables DeepSAVA to perform a very sparse attack on videos for maintaining human imperceptibility while still achieving state-of-the-art performance in terms of both attack success rate and adversarial transferability. Our intensive experiments on various types of deep neural networks and video datasets confirm the superiority of DeepSAVA.

Key facts. Cited by 11 papers in the map (data through 7 Oct 2026). Builds on 42 of its 65 references that are in the map. First posted on arXiv 10 Nov 2021.

Builds on

42 of 65 references.

References · 42

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  3. 2015Rethinking the Inception Architecture for Computer Visioncited by 5,451
  4. 2013Intriguing properties of neural networkscited by 5,114
  5. 2014Going Deeper with Convolutionscited by 4,231
  6. 2014Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translationcited by 4,016
  7. 2017Quo Vadis, Action Recognition? A New Model and the Kinetics Datasetcited by 3,259
  8. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
  9. 2015Spatial Transformer Networkscited by 1,423
  10. 2015DeepFool: a simple and accurate method to fool deep neural networkscited by 1,315
  11. 2014Deep Neural Networks are Easily Fooled: High Confidence Predictions for Unrecognizable Imagescited by 774
  12. 2016Delving into Transferable Adversarial Examples and Black-box Attackscited by 530
30 more
  1. 2014Long-term Recurrent Convolutional Networks for Visual Recognition and Descriptioncited by 483
  2. 2017Deep Learning for Medical Image Analysiscited by 399
  3. 2017Adversarial Examples: Attacks and Defenses for Deep Learningcited by 350
  4. 2018Generating Adversarial Examples with Adversarial Networkscited by 256
  5. 2018Human Action Recognition and Prediction: A Surveycited by 226
  6. 2018Spatially Transformed Adversarial Examplescited by 195
  7. 2020Perceptual Adversarial Robustness: Defense Against Unseen Threat Modelscited by 137
  8. 2018Reachability Analysis of Deep Neural Networks with Provable Guaranteescited by 93
  9. 2019Wasserstein Adversarial Examples via Projected Sinkhorn Iterationscited by 89
  10. 2019Functional Adversarial Attackscited by 82
  11. 2019Towards Large yet Imperceptible Adversarial Image Perturbations with Perceptual Color Distancecited by 72
  12. 2016A Boundary Tilting Persepective on the Phenomenon of Adversarial Examplescited by 70
  13. 2019Black-box Adversarial Attacks on Video Recognition Modelscited by 58
  14. 2018Sparse Adversarial Perturbations for Videoscited by 56
  15. 2018Concolic Testing for Deep Neural Networkscited by 51
  16. 2018Maximal Jacobian-based Saliency Map Attackcited by 46
  17. 2019Heuristic Black-box Adversarial Attacks on Video Recognition Modelscited by 37
  18. 2018A Game-Based Approximate Verification of Deep Neural Networks with Provable Guaranteescited by 33
  19. 2018Adversarial Perturbations Against Real-Time Video Classification Systemscited by 32
  20. 2020Sparse Black-box Video Attack with Reinforcement Learningcited by 24
  21. 2021Adversarial Driving: Attacking End-to-End Autonomous Drivingcited by 19
  22. 2019Appending Adversarial Frames for Universal Video Attackcited by 17
  23. 2020Generalizing Universal Adversarial Attacks Beyond Additive Perturbationscited by 17
  24. 2019Perceptual Evaluation of Adversarial Attacks for CNN-based Image Classificationcited by 12
  25. 2019Quantifying Perceptual Distortion of Adversarial Examplescited by 10
  26. 2019Perceptual Quality-preserving Black-Box Attack against Deep Learning Image Classifierscited by 8
  27. 2021Fooling Object Detectors: Adversarial Attacks by Half-Neighbor Maskscited by 5
  28. 2021Adversarial Robustness of Deep Learning: Theory, Algorithms, and Applicationscited by 5
  29. 2020Towards the Quantification of Safety Risks in Deep Neural Networkscited by 4
  30. 2021Tutorials on Testing Neural Networkscited by 2

Built on by

Cited by year

Citing papers per year: 2021: 1; 2022: 2; 2023: 3; 2024: 4; 2025: 1.

11 papers.

  1. 2022Enhancing Adversarial Training with Second-Order Statistics of Weightscited by 42
  2. 2023Efficient Decision-based Black-box Patch Attacks on Video Recognitioncited by 21
  3. 2023Towards Verifying the Geometric Robustness of Large-scale Neural Networkscited by 11
  4. 2024Boosting Adversarial Training via Fisher-Rao Norm-based Regularizationcited by 7
  5. 2023Reward Certification for Policy Smoothed Reinforcement Learningcited by 6
  6. 20223DVerifier: Efficient Robustness Verification for 3D Point Cloud Modelscited by 5
  7. 2024Seeing is Deceiving: Exploitation of Visual Pathways in Multi-Modal Language Modelscited by 4
  8. 2024Fine-Grained Alignment in Vision-and-Language Navigation through Bayesian Optimizationcited by 2
  9. 2021SNEAK: Synonymous Sentences-Aware Adversarial Attack on Natural Language Video Localizationcited by 1
  10. 2024SVASTIN: Sparse Video Adversarial Attack via Spatio-Temporal Invertible Neural Networkscited by 1
  11. 2025Adversarial Attacks in Multimodal Systems: A Practitioner's Surveycited by 0