mlsanityData to 7 Oct 2026

Attacking Motion Estimation with Adversarial Snow

arXiv 2210.11242 ↗First posted 20 Oct 2022cs.CV cs.LGcited by 9

Not on the map. Why?

Abstract (arXiv)

Current adversarial attacks for motion estimation (optical flow) optimize small per-pixel perturbations, which are unlikely to appear in the real world. In contrast, we exploit a real-world weather phenomenon for a novel attack with adversarially optimized snow. At the core of our attack is a differentiable renderer that consistently integrates photorealistic snowflakes with realistic motion into the 3D scene. Through optimization we obtain adversarial snow that significantly impacts the optical flow while being indistinguishable from ordinary snow. Surprisingly, the impact of our novel attack is largest on methods that previously showed a high robustness to small L_p perturbations.

Key facts. Cited by 9 papers in the map (data through 7 Oct 2026). Builds on 14 of its 17 references that are in the map. First posted on arXiv 20 Oct 2022.

Builds on

14 of 17 references.

References · 14

  1. 2019Benchmarking Neural Network Robustness to Common Corruptions and Perturbationscited by 2,857
  2. 2016FlowNet 2.0: Evolution of Optical Flow Estimation with Deep Networkscited by 780
  3. 2016Optical Flow Estimation using a Spatial Pyramid Networkcited by 421
  4. 2019Benchmarking Robustness in Object Detection: Autonomous Driving when Winter is Comingcited by 332
  5. 2021Learning to Estimate Hidden Motions with Global Motion Aggregationcited by 223
  6. 2020Learning perturbation sets for robust machine learningcited by 48
  7. 2019Attacking Optical Flowcited by 41
  8. 2021When Human Pose Estimation Meets Robustness: Adversarial Algorithms and Benchmarkscited by 41
  9. 2019Achieving Robustness in the Wild via Adversarial Mixing with Disentangled Representationscited by 28
  10. 2020Model-Based Robust Deep Learning: Generalizing to Natural, Out-of-Distribution Datacited by 27
  11. 2022A Perturbation-Constrained Adversarial Attack for Evaluating the Robustness of Optical Flowcited by 16
  12. 2021Towards Understanding Adversarial Robustness of Optical Flow Networkscited by 15
2 more
  1. 2022fakeWeather: Adversarial Attacks for Deep Neural Networks Emulating Weather Conditions on the Camera Lens of Autonomous Systemscited by 8
  2. 2020Deep Snow: Synthesizing Remote Sensing Imagery with Generative Adversarial Netscited by 5

Built on by

Cited by year

Citing papers per year: 2023: 4; 2024: 2; 2025: 3.

9 papers.

  1. 2023CosPGD: an efficient white-box adversarial attack for pixel-wise prediction taskscited by 27
  2. 2023Distracting Downpour: Adversarial Weather Attacks for Motion Estimationcited by 15
  3. 2023Improving Feature Stability during Upsampling -- Spectral Artifacts and the Importance of Spatial Contextcited by 8
  4. 2025RobustSpring: Benchmarking Robustness to Image Corruptions for Optical Flow, Scene Flow and Stereocited by 7
  5. 2023Detection Defenses: An Empty Promise against Adversarial Patch Attacks on Optical Flowcited by 6
  6. 2024Beware of Aliases -- Signal Preservation is Crucial for Robust Image Restorationcited by 6
  7. 2024Object segmentation from common fate: Motion energy processing enables human-like zero-shot generalization to random dot stimulicited by 5
  8. 2025Are Synthetic Corruptions A Reliable Proxy For Real-World Corruptions?cited by 5
  9. 2025DispBench: Benchmarking Disparity Estimation to Synthetic Corruptionscited by 0