mlsanityData to 7 Oct 2026

Towards Better Certified Segmentation via Diffusion Models

arXiv 2306.09949 ↗First posted 16 Jun 2023cs.CVcited by 2

On the map: Diffusion models → Medical imaging and segmentation

Abstract (arXiv)

The robustness of image segmentation has been an important research topic in the past few years as segmentation models have reached production-level accuracy. However, like classification models, segmentation models can be vulnerable to adversarial perturbations, which hinders their use in critical-decision systems like healthcare or autonomous driving. Recently, randomized smoothing has been proposed to certify segmentation predictions by adding Gaussian noise to the input to obtain theoretical guarantees. However, this method exhibits a trade-off between the amount of added noise and the level of certification achieved. In this paper, we address the problem of certifying segmentation prediction using a combination of randomized smoothing and diffusion models. Our experiments show that combining randomized smoothing and diffusion models significantly improves certified robustness, with results indicating a mean improvement of 21 points in accuracy compared to previous state-of-the-art methods on Pascal-Context and Cityscapes public datasets. Our method is independent of the selected segmentation model and does not need any additional specialized training procedure.

Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Placed in Medical imaging and segmentation, one of 14 directions in Diffusion models. Builds on 49 of its 58 references that are in the map. First posted on arXiv 16 Jun 2023.

Builds on

49 of 58 references.

References · 49

  1. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  2. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,185
  3. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  4. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,005
  5. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  6. 2022Generative Adversarial Networkscited by 5,603
  7. 2013Intriguing properties of neural networkscited by 5,114
  8. 2016The Cityscapes Dataset for Semantic Urban Scene Understandingcited by 4,427
  9. 2021Improved Denoising Diffusion Probabilistic Modelscited by 3,736
  10. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
  11. 2016Adversarial examples in the physical worldcited by 1,762
  12. 2018Spectral Normalization for Generative Adversarial Networkscited by 1,618
37 more
  1. 2020Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attackscited by 1,442
  2. 2019Deep High-Resolution Representation Learning for Visual Recognitioncited by 1,417
  3. 2019Certified Adversarial Robustness via Randomized Smoothingcited by 1,324
  4. 2018Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examplescited by 1,259
  5. 2022Vision Transformer Adapter for Dense Predictionscited by 564
  6. 2018Certified Robustness to Adversarial Examples with Differential Privacycited by 462
  7. 2019Provably Robust Deep Learning via Adversarially Trained Smoothed Classifierscited by 328
  8. 2018On the Effectiveness of Interval Bound Propagation for Training Verifiably Robust Modelscited by 322
  9. 2017Adversarial Examples for Semantic Segmentation and Object Detectioncited by 275
  10. 2018Lipschitz-Margin Training: Scalable Certification of Perturbation Invariance for Deep Neural Networkscited by 193
  11. 2018Certified Adversarial Robustness with Additive Noisecited by 178
  12. 2020Randomized Smoothing of All Shapes and Sizescited by 146
  13. 2018Generating 3D Adversarial Point Cloudscited by 144
  14. 2021Towards Robust Vision Transformercited by 137
  15. 2022(Certified!!) Adversarial Robustness for Free!cited by 130
  16. 2017On the Robustness of Semantic Segmentation Models to Adversarial Attackscited by 126
  17. 2021Orthogonalizing Convolutional Layers with the Cayley Transformcited by 105
  18. 2019Preventing Gradient Attenuation in Lipschitz Constrained Convolutional Networkscited by 86
  19. 2018Generalizable Adversarial Training via Spectral Normalizationcited by 68
  20. 2021Skew Orthogonal Convolutionscited by 67
  21. 2021Training Certifiably Robust Neural Networks with Efficient Local Lipschitz Boundscited by 64
  22. 2021A Dynamical System Perspective for Lipschitz Neural Networkscited by 59
  23. 2023A Unified Algebraic Perspective on Lipschitz Neural Networkscited by 58
  24. 2020Curse of Dimensionality on Randomized Smoothing for Certifiable Robustnesscited by 56
  25. 2021Mind the box: l1-APGD for sparse adversarial attacks on image classifierscited by 50
  26. 2022Almost-Orthogonal Layers for Efficient General-Purpose Lipschitz Networkscited by 39
  27. 2020Higher-Order Certification for Randomized Smoothingcited by 36
  28. 2020Random Smoothing Might be Unable to Certify ℓ_∞ Robustness for High-Dimensional Imagescited by 35
  29. 2021Scalable Certified Segmentation via Randomized Smoothingcited by 35
  30. 2019Theoretical evidence for adversarial robustness through randomizationcited by 33
  31. 2019Cubic LSTMs for Video Predictioncited by 32
  32. 2022LOT: Layer-wise Orthogonal Training on Improving ℓ2 Certified Robustnesscited by 32
  33. 2021Center Smoothing: Certified Robustness for Networks with Structured Outputscited by 18
  34. 2020Extensions and limitations of randomized smoothing for robustness guaranteescited by 15
  35. 2022Certified Defences Against Adversarial Patch Attacks on Semantic Segmentationcited by 15
  36. 2020Advocating for Multiple Defense Strategies against Adversarial Examplescited by 11
  37. 2022Towards Evading the Limits of Randomized Smoothing: A Theoretical Analysiscited by 2

Built on by

Cited by year

Citing papers per year: 2023: 1; 2024: 1.

2 papers.

  1. 2023Certification of Deep Learning Models for Medical Image Segmentationcited by 4
  2. 2024DiffYOLO: Object Detection for Anti-Noise via YOLO and Diffusion Modelscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2022DensePure: Understanding Diffusion Models towards Adversarial Robustness
  2. 2024Analysing Diffusion Segmentation for Medical Images
  3. 2025Robustifying Diffusion-Denoised Smoothing Against Covariate Shift
  4. 2025RDDPM: Robust Denoising Diffusion Probabilistic Model for Unsupervised Anomaly Segmentation
  5. 2025Robust Representation Consistency Model via Contrastive Denoising
  6. 2024Hybrid diffusion models: combining supervised and generative pretraining for label-efficient fine-tuning of segmentation models
  7. 2021Diffusion Models for Implicit Image Segmentation Ensembles
  8. 2023Importance of Aligning Training Strategy with Evaluation for Diffusion Models in 3D Multiclass Segmentation