mlsanityData to 14 Oct 2026

Beware of Aliases -- Signal Preservation is Crucial for Robust Image Restoration

arXiv 2406.07435 ↗First posted 11 Jun 2024cs.CV cs.LG eess.IVcited by 6

Not on the map. Why?

Abstract (arXiv)

Image restoration networks are usually comprised of an encoder and a decoder, responsible for aggregating image content from noisy, distorted data and to restore clean, undistorted images, respectively. Data aggregation as well as high-resolution image generation both usually come at the risk of involving aliases, i.e.~standard architectures put their ability to reconstruct the model input in jeopardy to reach high PSNR values on validation data. The price to be paid is low model robustness. In this work, we show that simply providing alias-free paths in state-of-the-art reconstruction transformers supports improved model robustness at low costs on the restoration performance. We do so by proposing BOA-Restormer, a transformer-based image restoration model that executes downsampling and upsampling operations partly in the frequency domain to ensure alias-free paths along the entire model while potentially preserving all relevant high-frequency information.

Key facts. Cited by 6 papers in the map (data through 14 Oct 2026). Builds on 39 of its 64 references that are in the map. First posted on arXiv 11 Jun 2024.

Builds on

39 of 64 references.

References · 39

  1. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
  2. 2014Explaining and Harnessing Adversarial Examplescited by 7,364
  3. 2022Generative Adversarial Networkscited by 5,627
  4. 2014Fully Convolutional Networks for Semantic Segmentationcited by 4,859
  5. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,998
  6. 2016Pyramid Scene Parsing Networkcited by 2,479
  7. 2015Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networkscited by 2,054
  8. 2020Shortcut Learning in Deep Neural Networkscited by 1,969
  9. 2021Restormer: Efficient Transformer for High-Resolution Image Restorationcited by 1,627
  10. 2019Theoretically Principled Trade-off between Robustness and Accuracycited by 1,517
  11. 2015SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentationcited by 1,487
  12. 2016Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Networkcited by 1,204
27 more
  1. 2021Alias-Free Generative Adversarial Networkscited by 1,184
  2. 2016Adversarial Machine Learning at Scalecited by 936
  3. 2018Robustness May Be at Odds with Accuracycited by 880
  4. 2016Deep Multi-scale Convolutional Neural Network for Dynamic Scene Deblurringcited by 727
  5. 2022Simple Baselines for Image Restorationcited by 708
  6. 2018Pyramid Stereo Matching Networkcited by 489
  7. 2019Making Convolutional Networks Shift-Invariant Againcited by 391
  8. 2015Learning Deconvolution Network for Semantic Segmentationcited by 365
  9. 2016Deep Joint Rain Detection and Removal from a Single Imagecited by 352
  10. 2020Watch your Up-Convolution: CNN Based Generative Deep Neural Networks are Failing to Reproduce Spectral Distributionscited by 280
  11. 2021Global Filter Networks for Image Classificationcited by 260
  12. 2017Image De-raining Using a Conditional Generative Adversarial Networkcited by 197
  13. 2021Intriguing Findings of Frequency Selection for Image Deblurringcited by 81
  14. 2022SegPGD: An Effective and Efficient Adversarial Attack for Evaluating and Boosting Segmentation Robustnesscited by 74
  15. 2020Delving Deeper into Anti-aliasing in ConvNetscited by 60
  16. 2020Spectral Distribution Aware Image Generationcited by 29
  17. 2021WaveCNet: Wavelet Integrated CNNs to Suppress Aliasing Effect for Noise-Robust Image Classificationcited by 29
  18. 2022FrequencyLowCut Pooling -- Plug & Play against Catastrophic Overfittingcited by 28
  19. 2023CosPGD: an efficient white-box adversarial attack for pixel-wise prediction taskscited by 27
  20. 2022A Perturbation-Constrained Adversarial Attack for Evaluating the Robustness of Optical Flowcited by 16
  21. 2021Towards Understanding Adversarial Robustness of Optical Flow Networkscited by 15
  22. 2021Anti-aliasing Deep Image Classifiers using Novel Depth Adaptive Blurring and Activation Functioncited by 11
  23. 2023On the unreasonable vulnerability of transformers for image restoration -- and an easy fixcited by 11
  24. 2022Attacking Motion Estimation with Adversarial Snowcited by 9
  25. 2023As large as it gets: Learning infinitely large Filters via Neural Implicit Functions in the Fourier Domaincited by 8
  26. 2023Differentiable Sensor Layouts for End-to-End Learning of Task-Specific Camera Parameterscited by 6
  27. 2022Periodic Artifact Reduction in Fourier transforms of Full Field Atomic Resolution Imagescited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 1; 2025: 4; 2026: 1.

6 papers.

  1. 2025RobustSpring: Benchmarking Robustness to Image Corruptions for Optical Flow, Scene Flow and Stereocited by 7
  2. 2025Are Synthetic Corruptions A Reliable Proxy For Real-World Corruptions?cited by 5
  3. 2024How Do Training Methods Influence the Utilization of Vision Models?cited by 3
  4. 2025Smart Eyes for Silent Threats: VLMs and In-Context Learning for THz Imagingcited by 2
  5. 2025DispBench: Benchmarking Disparity Estimation to Synthetic Corruptionscited by 0
  6. 2026Failure or Drift? Evaluating Monocular SLAM under Synthetic and Real-World Corruptionscited by 0