mlsanityData to 7 Oct 2026

Uncertainty-weighted Loss Functions for Improved Adversarial Attacks on Semantic Segmentation

arXiv 2310.17436 ↗First posted 26 Oct 2023cs.CVcited by 9

Not on the map. Why?

Abstract (arXiv)

State-of-the-art deep neural networks have been shown to be extremely powerful in a variety of perceptual tasks like semantic segmentation. However, these networks are vulnerable to adversarial perturbations of the input which are imperceptible for humans but lead to incorrect predictions. Treating image segmentation as a sum of pixel-wise classifications, adversarial attacks developed for classification models were shown to be applicable to segmentation models as well. In this work, we present simple uncertainty-based weighting schemes for the loss functions of such attacks that (i) put higher weights on pixel classifications which can more easily perturbed and (ii) zero-out the pixel-wise losses corresponding to those pixels that are already confidently misclassified. The weighting schemes can be easily integrated into the loss function of a range of well-known adversarial attackers with minimal additional computational overhead, but lead to significant improved perturbation performance, as we demonstrate in our empirical analysis on several datasets and models.

Key facts. Cited by 9 papers in the map (data through 7 Oct 2026). Builds on 30 of its 34 references that are in the map. First posted on arXiv 26 Oct 2023.

Builds on

30 of 34 references.

References · 30

  1. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  2. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  3. 2016The Cityscapes Dataset for Semantic Urban Scene Understandingcited by 4,427
  4. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,416
  5. 2018Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentationcited by 3,210
  6. 2016Pyramid Scene Parsing Networkcited by 2,472
  7. 2015DeepFool: a simple and accurate method to fool deep neural networkscited by 1,315
  8. 2016Adversarial Machine Learning at Scalecited by 934
  9. 2018BiSeNet: Bilateral Segmentation Network for Real-time Semantic Segmentationcited by 645
  10. 2017Adversarial Examples for Semantic Segmentation and Object Detectioncited by 275
  11. 2017On the Robustness of Semantic Segmentation Models to Adversarial Attackscited by 126
  12. 2017Universal Adversarial Perturbations Against Semantic Image Segmentationcited by 92
18 more
  1. 2022SegPGD: An Effective and Efficient Adversarial Attack for Evaluating and Boosting Segmentation Robustnesscited by 74
  2. 2021Evaluating the Robustness of Semantic Segmentation for Autonomous Driving against Real-World Adversarial Patch Attackscited by 55
  3. 2017Houdini: Fooling Deep Structured Prediction Modelscited by 50
  4. 2018Characterizing Adversarial Examples Based on Spatial Consistency Information for Semantic Segmentationcited by 40
  5. 2020Dynamic Divide-and-Conquer Adversarial Training for Robust Semantic Segmentationcited by 38
  6. 2021Evading Adversarial Example Detection Defenses with Orthogonal Projected Gradient Descentcited by 34
  7. 2023CosPGD: an efficient white-box adversarial attack for pixel-wise prediction taskscited by 27
  8. 2020Improved Noise and Attack Robustness for Semantic Segmentation by Using Multi-Task Training with Self-Supervised Depth Estimationcited by 26
  9. 2019Indirect Local Attacks for Context-aware Semantic Segmentation Networkscited by 23
  10. 2022Proximal Splitting Adversarial Attacks for Semantic Segmentationcited by 23
  11. 2019Time-Dynamic Estimates of the Reliability of Deep Semantic Segmentation Networkscited by 20
  12. 2022Certified Defences Against Adversarial Patch Attacks on Semantic Segmentationcited by 15
  13. 2019DAPAS : Denoising Autoencoder to Prevent Adversarial attack in Semantic Segmentationcited by 14
  14. 2021The Vulnerability of Semantic Segmentation Networks to Adversarial Attacks in Autonomous Driving: Enhancing Extensive Environment Sensingcited by 12
  15. 2023Pixel-wise Gradient Uncertainty for Convolutional Neural Networks applied to Out-of-Distribution Segmentationcited by 11
  16. 2021Semantically Stealthy Adversarial Attacks against Segmentation Modelscited by 7
  17. 2023A Certified Radius-Guided Attack Framework to Image Segmentation Modelscited by 5
  18. 2023Uncertainty-based Detection of Adversarial Attacks in Semantic Segmentationcited by 5

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 3; 2025: 5; 2026: 1.

9 papers.

  1. 2024Reducing Texture Bias of Deep Neural Networks via Edge Enhancing Diffusioncited by 11
  2. 2025SegTrans: Transferable Adversarial Examples for Segmentation Modelscited by 8
  3. 2024Uni-SLAM: Uncertainty-Aware Neural Implicit SLAM for Real-Time Dense Indoor Scene Reconstructioncited by 3
  4. 2025Semantic Representation Attack against Aligned Large Language Modelscited by 3
  5. 2025Bridging the Task Gap: Multi-Task Adversarial Transferability in CLIP and Its Derivativescited by 2
  6. 2024Detecting Adversarial Attacks in Semantic Segmentation via Uncertainty Estimation: A Deep Analysiscited by 1
  7. 2025Transferring Styles for Reduced Texture Bias and Improved Robustness in Semantic Segmentation Networkscited by 1
  8. 2025Proxy-Embedding as an Adversarial Teacher: An Embedding-Guided Bidirectional Attack for Referring Expression Segmentation Modelscited by 0
  9. 2026IGME: Efficient Chained Method Ensemble for Transferable Semantic Segmentation Attackscited by 0