mlsanityData to 7 Oct 2026

Making Convolutional Networks Shift-Invariant Again

arXiv 1904.11486 ↗First posted 25 Apr 2019cs.CV cs.LGcited by 390

Not on the map. Why?

Cited by 23 in Diffusion models, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

Modern convolutional networks are not shift-invariant, as small input shifts or translations can cause drastic changes in the output. Commonly used downsampling methods, such as max-pooling, strided-convolution, and average-pooling, ignore the sampling theorem. The well-known signal processing fix is anti-aliasing by low-pass filtering before downsampling. However, simply inserting this module into deep networks degrades performance; as a result, it is seldomly used today. We show that when integrated correctly, it is compatible with existing architectural components, such as max-pooling and strided-convolution. We observe textitincreased accuracy in ImageNet classification, across several commonly-used architectures, such as ResNet, DenseNet, and MobileNet, indicating effective regularization. Furthermore, we observe textitbetter generalization, in terms of stability and robustness to input corruptions. Our results demonstrate that this classical signal processing technique has been undeservingly overlooked in modern deep networks. Code and anti-aliased versions of popular networks are available at https://richzhang.github.io/antialiased-cnns/ .

Key facts. Cited by 390 papers in the map (data through 7 Oct 2026). First posted on arXiv 25 Apr 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 107; 2022: 79; 2023: 72; 2024: 43; 2025: 54; 2026: 35.

390 papers.

Sort:

By area:

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  2. 2021Mip-NeRF: A Multiscale Representation for Anti-Aliasing Neural Radiance Fieldscited by 1,957
  3. 2021Alias-Free Generative Adversarial Networkscited by 1,181
  4. 2023Vision-Language Models for Vision Tasks: A Surveycited by 834
  5. 2024Emu3: Next-Token Prediction is All You Needcited by 761
  6. 2021Score-based Generative Modeling in Latent Spacecited by 634
  7. 2023Language Model Beats Diffusion -- Tokenizer is Key to Visual Generationcited by 633
  8. 2021Tackling the Generative Learning Trilemma with Denoising Diffusion GANscited by 515
  9. 2021StyleNeRF: A Style-based 3D-Aware Generator for High-resolution Image Synthesiscited by 477
  10. 2021Deep ViT Features as Dense Visual Descriptorscited by 327
  11. 2023DreamSim: Learning New Dimensions of Human Visual Similarity using Synthetic Datacited by 322
  12. 2021Score-based diffusion models for accelerated MRIcited by 308
38 more
  1. 2022How Do Vision Transformers Work?cited by 305
  2. 2021Score-Based Generative Modeling with Critically-Damped Langevin Diffusioncited by 230
  3. 2021HeadNeRF: A Real-time NeRF-based Parametric Head Modelcited by 228
  4. 2021Scaling Local Self-Attention for Parameter Efficient Visual Backbonescited by 220
  5. 2021Towards Causal Representation Learningcited by 207
  6. 2022Speech Enhancement and Dereverberation with Diffusion-based Generative Modelscited by 195
  7. 2025GAIA-2: A Controllable Multi-View Generative World Model for Autonomous Drivingcited by 178
  8. 2021Revisiting ResNets: Improved Training and Scaling Strategiescited by 158
  9. 2021Towards Robust Vision Transformercited by 137
  10. 2021Unconstrained Scene Generation with Locally Conditioned Radiance Fieldscited by 130
  11. 2022Photorealistic Monocular 3D Reconstruction of Humans Wearing Clothingcited by 126
  12. 2021AdaMatch: A Unified Approach to Semi-Supervised Learning and Domain Adaptationcited by 104
  13. 2021Fast Point Transformercited by 103
  14. 2022RelPose: Predicting Probabilistic Relative Rotation for Single Objects in the Wildcited by 102
  15. 2021Involution: Inverting the Inherence of Convolution for Visual Recognitioncited by 99
  16. 2024MaskBit: Embedding-free Image Generation via Bit Tokenscited by 96
  17. 2021AugMax: Adversarial Composition of Random Augmentations for Robust Trainingcited by 89
  18. 2023MedViT: A Robust Vision Transformer for Generalized Medical Image Classificationcited by 89
  19. 2021Diverse Branch Block: Building a Convolution as an Inception-like Unitcited by 87
  20. 2023RelPose++: Recovering 6D Poses from Sparse-view Observationscited by 87
  21. 2021Ensembling Off-the-shelf Models for GAN Trainingcited by 84
  22. 2022Wave-ViT: Unifying Wavelet and Transformers for Visual Representation Learningcited by 78
  23. 2021Multimodal Conditional Image Synthesis with Product-of-Experts GANscited by 75
  24. 2022VCT: A Video Compression Transformercited by 74
  25. 2021Aligning Latent and Image Spaces to Connect the Unconnectablecited by 70
  26. 2021LEAF: A Learnable Frontend for Audio Classificationcited by 69
  27. 2021FlexConv: Continuous Kernel Convolutions with Differentiable Kernel Sizescited by 66
  28. 2024DINO-Tracker: Taming DINO for Self-Supervised Point Tracking in a Single Videocited by 64
  29. 2021On the Frequency Bias of Generative Modelscited by 60
  30. 2024FouriScale: A Frequency Perspective on Training-Free High-Resolution Image Synthesiscited by 58
  31. 2021Adversarial Robustness Comparison of Vision Transformer and MLP-Mixer to CNNscited by 57
  32. 2023Towards Stability of Autoregressive Neural Operatorscited by 54
  33. 2021Ensembling with Deep Generative Viewscited by 53
  34. 2021On Aliased Resizing and Surprising Subtleties in GAN Evaluationcited by 51
  35. 2022The Lie Derivative for Measuring Learned Equivariancecited by 51
  36. 2023FFCV: Accelerating Training by Removing Data Bottleneckscited by 49
  37. 2021A Systematic Review of Robustness in Deep Learning for Computer Vision: Mind the gap?cited by 48
  38. 2024Frequency-aware Feature Fusion for Dense Image Predictioncited by 48

50 most cited of 390 All 390 (JSON)