mlsanityData to 7 Oct 2026

Perceptual Attacks of No-Reference Image Quality Models with Human-in-the-Loop

arXiv 2210.00933 ↗First posted 3 Oct 2022cs.CV eess.IVcited by 30

Not on the map. Why?

Abstract (arXiv)

No-reference image quality assessment (NR-IQA) aims to quantify how humans perceive visual distortions of digital images without access to their undistorted references. NR-IQA models are extensively studied in computational vision, and are widely used for performance evaluation and perceptual optimization of man-made vision systems. Here we make one of the first attempts to examine the perceptual robustness of NR-IQA models. Under a Lagrangian formulation, we identify insightful connections of the proposed perceptual attack to previous beautiful ideas in computer vision and machine learning. We test one knowledge-driven and three data-driven NR-IQA methods under four full-reference IQA models (as approximations to human perception of just-noticeable differences). Through carefully designed psychophysical experiments, we find that all four NR-IQA models are vulnerable to the proposed perceptual attack. More interestingly, we observe that the generated counterexamples are not transferable, manifesting themselves as distinct design flows of respective NR-IQA methods.

Key facts. Cited by 30 papers in the map (data through 7 Oct 2026). Builds on 33 of its 71 references that are in the map. First posted on arXiv 3 Oct 2022.

Builds on

33 of 71 references.

References · 33

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  3. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,029
  4. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  5. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  6. 2013Intriguing properties of neural networkscited by 5,114
  7. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
  8. 2016Adversarial examples in the physical worldcited by 1,762
  9. 2019Theoretically Principled Trade-off between Robustness and Accuracycited by 1,510
  10. 2019Do ImageNet Classifiers Generalize to ImageNet?cited by 1,507
  11. 2015DeepFool: a simple and accurate method to fool deep neural networkscited by 1,315
  12. 2020Image Quality Assessment: Unifying Structure and Texture Similaritycited by 792
21 more
  1. 2014Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognitioncited by 719
  2. 2017Evasion Attacks against Machine Learning at Test Timecited by 686
  3. 2016End-to-end Optimized Image Compressioncited by 673
  4. 2019On Evaluating Adversarial Robustnesscited by 442
  5. 2019KonIQ-10k: An ecologically valid database for deep learning of blind image quality assessmentcited by 351
  6. 2017Adversarial Examples for Semantic Segmentation and Object Detectioncited by 275
  7. 2019From Patches to Pictures (PaQ-2-PiQ): Mapping the Perceptual Space of Picture Qualitycited by 221
  8. 2015Massive Online Crowdsourced Study of Subjective and Objective Picture Qualitycited by 208
  9. 2019Efficient Decision-based Black-box Adversarial Attacks on Face Recognitioncited by 173
  10. 2020Perceptual Adversarial Robustness: Defense Against Unseen Threat Modelscited by 137
  11. 2018Constructing Unrestricted Adversarial Examples with Generative Modelscited by 135
  12. 2020Uncertainty-Aware Blind Image Quality Assessment in the Laboratory and Wildcited by 134
  13. 2016Perceptual Quality Prediction on Authentically Distorted Images Using a Bag of Features Approachcited by 71
  14. 2021When Human Pose Estimation Meets Robustness: Adversarial Algorithms and Benchmarkscited by 41
  15. 2018Characterizing Adversarial Examples Based on Spatial Consistency Information for Semantic Segmentationcited by 40
  16. 2021Continual Learning for Blind Image Quality Assessmentcited by 36
  17. 2017VideoSet: A Large-Scale Compressed Video Quality Dataset Based on JND Measurementcited by 33
  18. 2020Robust Tracking against Adversarial Attackscited by 28
  19. 2017Eigen-Distortions of Hierarchical Representationscited by 25
  20. 2021Understanding the Robustness of Skeleton-based Action Recognition under Adversarial Attackcited by 23
  21. 2021Troubleshooting Blind Image Quality Models in the Wildcited by 14

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 1; 2023: 6; 2024: 15; 2025: 6; 2026: 2.

30 papers.

Sort:

  1. 2023Blind Image Quality Assessment via Vision-Language Correspondence: A Multitask Learning Perspectivecited by 276
  2. 2023Comparing the Robustness of Modern No-Reference Image- and Video-Quality Metrics to Adversarial Attackscited by 14
  3. 2024Exploring Rich Subjective Quality Information for Image Quality Assessment in the Wildcited by 13
  4. 2023Vulnerabilities in Video Quality Assessment Models: The Challenge of Adversarial Attackscited by 12
  5. 2024Backdoor Attacks against No-Reference Image Quality Assessment Models via a Scalable Triggercited by 12
  6. 2024Exploring Vulnerabilities of No-Reference Image Quality Assessment Models: A Query-Based Black-Box Methodcited by 10
  7. 2023Attacking Perceptual Similarity Metricscited by 8
  8. 2024Black-box Adversarial Attacks Against Image Quality Assessment Modelscited by 8
  9. 2025Cross-Modal Transferable Image-to-Video Attack on Video Quality Metricscited by 8
  10. 2023Fast Adversarial CNN-based Perturbation Attack on No-Reference Image- and Video-Quality Metricscited by 7
  11. 2024Defense Against Adversarial Attacks on No-Reference Image Quality Models with Gradient Norm Regularizationcited by 7
  12. 2024Adversarial purification for no-reference image-quality metrics: applicability study and new methodscited by 5
18 more
  1. 2024Guardians of Image Quality: Benchmarking Defenses Against Adversarial Attacks on Image Quality Metricscited by 5
  2. 2024IOI: Invisible One-Iteration Adversarial Attack on No-Reference Image- and Video-Quality Metricscited by 4
  3. 2024Ti-Patch: Tiled Physical Adversarial Patch for no-reference video quality metricscited by 4
  4. 2024Beyond Score Changes: Adversarial Attack on No-Reference Image Quality Assessment from Two Perspectivescited by 3
  5. 2024Quality Prediction of AI Generated Images and Videos: Emerging Trends and Opportunitiescited by 3
  6. 2025Test-Time Preference Optimization for Image Restorationcited by 2
  7. 2022Applicability limitations of differentiable full-reference image-qualitycited by 1
  8. 2024Secure Video Quality Assessment Resisting Adversarial Attackscited by 1
  9. 2025CleanMAP: Distilling Multimodal LLMs for Confidence-Driven Crowdsourced HD Map Updatescited by 1
  10. 2025SEGA: A Transferable Signed Ensemble Gaussian Black-Box Attack against No-Reference Image Quality Assessment Modelscited by 1
  11. 2026BiRQA: Bidirectional Robust Quality Assessment for Imagescited by 1
  12. 2023Learning with Noisy Low-Cost MOS for Image Quality Assessment via Dual-Bias Calibrationcited by 0
  13. 2024Causal Perception Inspired Representation Learning for Trustworthy Image Quality Assessmentcited by 0
  14. 2024Scale Contrastive Learning with Selective Attentions for Blind Image Quality Assessmentcited by 0
  15. 2024Stochastic BIQA: Median Randomized Smoothing for Certified Blind Image Quality Assessmentcited by 0
  16. 2025Robustness as Architecture: Designing IQA Models to Withstand Adversarial Perturbationscited by 0
  17. 2025FS-IQA: Certified Feature Smoothing for Robust Image Quality Assessmentcited by 0
  18. 2026Causal Disentanglement-Inspired Degradation Representation Learning for Full-Reference Image Quality Assessmentcited by 0