mlsanityData to 7 Oct 2026

Black-box Adversarial Attacks Against Image Quality Assessment Models

arXiv 2402.17533 ↗First posted 27 Feb 2024cs.CV eess.IVcited by 8

Not on the map. Why?

Abstract (arXiv)

The goal of No-Reference Image Quality Assessment (NR-IQA) is to predict the perceptual quality of an image in line with its subjective evaluation. To put the NR-IQA models into practice, it is essential to study their potential loopholes for model refinement. This paper makes the first attempt to explore the black-box adversarial attacks on NR-IQA models. Specifically, we first formulate the attack problem as maximizing the deviation between the estimated quality scores of original and perturbed images, while restricting the perturbed image distortions for visual quality preservation. Under such formulation, we then design a Bi-directional loss function to mislead the estimated quality scores of adversarial examples towards an opposite direction with maximum deviation. On this basis, we finally develop an efficient and effective black-box attack method against NR-IQA models. Extensive experiments reveal that all the evaluated NR-IQA models are vulnerable to the proposed attack method. And the generated perturbations are not transferable, enabling them to serve the investigation of specialities of disparate IQA models.

Key facts. Cited by 8 papers in the map (data through 7 Oct 2026). Builds on 27 of its 55 references that are in the map. First posted on arXiv 27 Feb 2024.

Builds on

27 of 55 references.

References · 27

  1. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  2. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,029
  3. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  4. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  5. 2013Intriguing properties of neural networkscited by 5,114
  6. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
  7. 2016Adversarial examples in the physical worldcited by 1,762
  8. 2020Image Quality Assessment: Unifying Structure and Texture Similaritycited by 792
  9. 2015The Limitations of Deep Learning in Adversarial Settingscited by 784
  10. 2019Square Attack: a query-efficient black-box adversarial attack via random searchcited by 682
  11. 2022MANIQA: Multi-dimension Attention Network for No-Reference Image Quality Assessmentcited by 514
  12. 2019KonIQ-10k: An ecologically valid database for deep learning of blind image quality assessmentcited by 351
15 more
  1. 2019Blind Image Quality Assessment Using A Deep Bilinear Convolutional Neural Networkcited by 301
  2. 2023Blind Image Quality Assessment via Vision-Language Correspondence: A Multitask Learning Perspectivecited by 276
  3. 2016Deep Neural Networks for No-Reference and Full-Reference Image Quality Assessmentcited by 224
  4. 2015Massive Online Crowdsourced Study of Subjective and Objective Picture Qualitycited by 208
  5. 2021No-Reference Image Quality Assessment via Transformers, Relative Ranking, and Self-Consistencycited by 165
  6. 2020MetaIQA: Deep Meta-learning for No-Reference Image Quality Assessmentcited by 144
  7. 2020Uncertainty-Aware Blind Image Quality Assessment in the Laboratory and Wildcited by 134
  8. 2016On the Use of Deep Learning for Blind Image Quality Assessmentcited by 34
  9. 2022Perceptual Attacks of No-Reference Image Quality Models with Human-in-the-Loopcited by 30
  10. 2022Universal Perturbation Attack on Differentiable No-Reference Image- and Video-Quality Metricscited by 17
  11. 2021Training Meta-Surrogate Model for Transferable Adversarial Attackcited by 15
  12. 2023Vulnerabilities in Video Quality Assessment Models: The Challenge of Adversarial Attackscited by 12
  13. 2024Exploring Vulnerabilities of No-Reference Image Quality Assessment Models: A Query-Based Black-Box Methodcited by 10
  14. 2023State-of-the-art optical-based physical adversarial attacks for deep learning computer vision systemscited by 8
  15. 2020A multiple attributes image quality database for smartphone camera photo quality assessmentcited by 3

Built on by

Cited by year

Citing papers per year: 2024: 3; 2025: 5.

8 papers.

  1. 2024Guardians of Image Quality: Benchmarking Defenses Against Adversarial Attacks on Image Quality Metricscited by 5
  2. 2024Ti-Patch: Tiled Physical Adversarial Patch for no-reference video quality metricscited by 4
  3. 2025AdvReal: Physical Adversarial Patch Generation Framework for Security Evaluation of Object Detection Systemscited by 4
  4. 2024Secure Video Quality Assessment Resisting Adversarial Attackscited by 1
  5. 2025TenAd: A Tensor-based Low-rank Black Box Adversarial Attack for Video Classificationcited by 1
  6. 2025SEGA: A Transferable Signed Ensemble Gaussian Black-Box Attack against No-Reference Image Quality Assessment Modelscited by 1
  7. 2025Adversarial Stress Testing of Outlier Detection in Subjective Image Quality Assessmentcited by 0
  8. 2025FeatureFool: Zero-Query Fooling of Video Models via Feature Mapcited by 0