Exploring adversarial robustness of JPEG AI: methodology, comparison and new methods
Not on the map. Why?
Abstract (arXiv)
Adversarial robustness of neural networks is an increasingly important area of research, combining studies on computer vision models, large language models (LLMs), and others. With the release of JPEG AI - the first standard for end-to-end neural image compression (NIC) methods - the question of its robustness has become critically significant. JPEG AI is among the first international, real-world applications of neural-network-based models to be embedded in consumer devices. However, research on NIC robustness has been limited to open-source codecs and a narrow range of attacks. This paper proposes a new methodology for measuring NIC robustness to adversarial attacks. We present the first large-scale evaluation of JPEG AI's robustness, comparing it with other NIC models. Our evaluation results and code are publicly available online (link is hidden for a blind review).
Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 29 of its 42 references that are in the map. First posted on arXiv 18 Nov 2024.
Builds on
29 of 42 references.
References · 29
- 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
- 2016The Cityscapes Dataset for Semantic Urban Scene Understandingcited by 4,444
- 2016Towards Evaluating the Robustness of Neural Networkscited by 2,998
- 2016Adversarial examples in the physical worldcited by 1,769
- 2018Variational image compression with a scale hyperpriorcited by 935
- 2016End-to-end Optimized Image Compressioncited by 696
- 2018Joint Autoregressive and Hierarchical Priors for Learned Image Compressioncited by 664
- 2020Learned Image Compression with Discretized Gaussian Mixture Likelihoods and Attention Modulescited by 590
- 2022Diffusion Models for Adversarial Purificationcited by 476
- 2020High-Fidelity Generative Image Compressioncited by 338
- 2022ELIC: Efficient Learned Image Compression with Unevenly Grouped Space-Channel Contextual Adaptive Codingcited by 301
- 2021Deep Contextual Video Compressioncited by 221
17 more
- 2023Learned Image Compression with Mixed Transformer-CNN Architecturescited by 211
- 2022Lossy Image Compression with Conditional Diffusion Modelscited by 170
- 2017Soft-to-Hard Vector Quantization for End-to-End Learning Compressible Representationscited by 152
- 2022The Devil Is in the Details: Window-based Attention for Image Compressioncited by 147
- 2019Unrestricted Adversarial Examples via Semantic Manipulationcited by 84
- 2020Improving Inference for Neural Image Compressioncited by 83
- 2022Frequency-driven Imperceptible Adversarial Attack on Semantic Similaritycited by 77
- 2022Physical Adversarial Attack meets Computer Vision: A Decade Surveycited by 67
- 2021A Survey On Universal Adversarial Attackcited by 59
- 2023EVC: Towards Real-Time Neural Image Compression with Mask Decaycited by 49
- 2021Slimmable Compressive Autoencoders for Practical Neural Image Compressioncited by 39
- 2022Lossy Image Compression with Quantized Hierarchical VAEscited by 35
- 2022Video compression dataset and benchmark of learning-based video-quality metricscited by 27
- 2021Towards Robust Neural Image Compression: Adversarial Attack and Model Finetuningcited by 25
- 2023Comparing the Robustness of Modern No-Reference Image- and Video-Quality Metrics to Adversarial Attackscited by 14
- 2020Pick-Object-Attack: Type-Specific Adversarial Attack for Object Detectioncited by 8
- 2024Attack and Defense Analysis of Learned Image Compressioncited by 3
Built on by
Citing papers per year: 2025: 3.
3 papers.