mlsanityData to 7 Oct 2026

3D Common Corruptions and Data Augmentation

arXiv 2203.01441 ↗First posted 2 Mar 2022cs.CV cs.LGcited by 120

Not on the map. Why?

Abstract (arXiv)

We introduce a set of image transformations that can be used as corruptions to evaluate the robustness of models as well as data augmentation mechanisms for training neural networks. The primary distinction of the proposed transformations is that, unlike existing approaches such as Common Corruptions, the geometry of the scene is incorporated in the transformations -- thus leading to corruptions that are more likely to occur in the real world. We also introduce a set of semantic corruptions (e.g. natural object occlusions). We show these transformations are `efficient' (can be computed on-the-fly), `extendable' (can be applied on most image datasets), expose vulnerability of existing models, and can effectively make models more robust when employed as `3D data augmentation' mechanisms. The evaluations on several tasks and datasets suggest incorporating 3D information into benchmarking and training opens up a promising direction for robustness research.

Key facts. Cited by 120 papers in the map (data through 7 Oct 2026). Builds on 65 of its 85 references that are in the map. First posted on arXiv 2 Mar 2022.

Builds on

65 of 85 references.

References · 65

  1. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  2. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  3. 2014Microsoft COCO: Common Objects in Contextcited by 17,498
  4. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  5. 2017mixup: Beyond Empirical Risk Minimizationcited by 5,155
  6. 2013Intriguing properties of neural networkscited by 5,114
  7. 2019CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Featurescited by 2,875
  8. 2019Benchmarking Neural Network Robustness to Common Corruptions and Perturbationscited by 2,857
  9. 2020Shortcut Learning in Deep Neural Networkscited by 1,928
  10. 2020The Many Faces of Robustness: A Critical Analysis of Out-of-Distribution Generalizationcited by 1,842
  11. 2021Vision Transformers for Dense Predictioncited by 1,789
  12. 2017Improved Regularization of Convolutional Neural Networks with Cutoutcited by 1,623
38 more
  1. 2019Towards Robust Monocular Depth Estimation: Mixing Datasets for Zero-shot Cross-dataset Transfercited by 1,484
  2. 2018ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustnesscited by 1,430
  3. 2016Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoisingcited by 1,352
  4. 2019Self-training with Noisy Student improves ImageNet classificationcited by 1,304
  5. 2016Adversarial Machine Learning at Scalecited by 934
  6. 2019AugMix: A Simple Data Processing Method to Improve Robustness and Uncertaintycited by 897
  7. 2019The Replica Dataset: A Digital Replica of Indoor Spacescited by 885
  8. 2019On the Convergence of Adam and Beyondcited by 834
  9. 2016Deep Multi-scale Convolutional Neural Network for Dynamic Scene Deblurringcited by 722
  10. 2020RobustBench: a standardized adversarial robustness benchmarkcited by 659
  11. 2018Taskonomy: Disentangling Task Transfer Learningcited by 631
  12. 2020Hypersim: A Photorealistic Synthetic Dataset for Holistic Indoor Scene Understandingcited by 593
  13. 2018Learning to See in the Darkcited by 459
  14. 2020Measuring Robustness to Natural Distribution Shifts in Image Classificationcited by 452
  15. 2021Intriguing Properties of Vision Transformerscited by 446
  16. 2017Learning Deep CNN Denoiser Prior for Image Restorationcited by 407
  17. 2017Semantic Foggy Scene Understanding with Synthetic Datacited by 401
  18. 2019Using Pre-Training Can Improve Model Robustness and Uncertaintycited by 373
  19. 2021Omnidata: A Scalable Pipeline for Making Multi-Task Mid-Level Vision Datasets from 3D Scanscited by 336
  20. 2019Benchmarking Robustness in Object Detection: Autonomous Driving when Winter is Comingcited by 332
  21. 20193D Scene Graph: A Structure for Unified Semantics, 3D Space, and Cameracited by 280
  22. 2020Data and its (dis)contents: A survey of dataset development and use in machine learning researchcited by 259
  23. 2021Understanding Robustness of Transformers for Image Classificationcited by 250
  24. 2021Accuracy on the Line: On the Strong Correlation Between Out-of-Distribution and In-Distribution Generalizationcited by 248
  25. 2019Improving Adversarial Robustness via Promoting Ensemble Diversitycited by 213
  26. 2020A Physics-based Noise Formation Model for Extreme Low-light Raw Denoisingcited by 138
  27. 20193D Ken Burns Effect from a Single Imagecited by 128
  28. 2019Adversarial Examples Are a Natural Consequence of Test Error in Noisecited by 114
  29. 2019Benchmarking the Robustness of Semantic Segmentation Modelscited by 111
  30. 2017A Study and Comparison of Human and Deep Learning Recognition Performance Under Visual Distortionscited by 107
  31. 2019Improving Robustness Without Sacrificing Accuracy with Patch Gaussian Augmentationcited by 105
  32. 2020Robust Learning Through Cross-Task Consistencycited by 105
  33. 2021On the Adversarial Robustness of Vision Transformerscited by 104
  34. 2020On Robustness and Transferability of Convolutional Neural Networkscited by 98
  35. 2021On Interaction Between Augmentations and Corruptions in Natural Corruption Robustnesscited by 98
  36. 2017Measuring the tendency of CNNs to Learn Surface Statistical Regularitiescited by 87
  37. 2022Benchmarking Robustness of 3D Point Cloud Recognition Against Common Corruptionscited by 78
  38. 2020DVERGE: Diversifying Vulnerabilities for Enhanced Robust Generation of Ensemblescited by 73

50 most cited of 65 All 65 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 6; 2023: 36; 2024: 33; 2025: 27; 2026: 18.

120 papers.

Sort:

By area:

  1. 2024Metric3Dv2: A Versatile Monocular Geometric Foundation Model for Zero-shot Metric Depth and Surface Normal Estimationcited by 389
  2. 2024GeoWizard: Unleashing the Diffusion Priors for 3D Geometry Estimation from a Single Imagecited by 283
  3. 2022Surgical Fine-Tuning Improves Adaptation to Distribution Shiftscited by 205
  4. 2024Lotus: Diffusion-based Visual Foundation Model for High-quality Dense Predictioncited by 163
  5. 2023Robo3D: Towards Robust and Reliable 3D Perception against Corruptionscited by 128
  6. 2024Rethinking Inductive Biases for Surface Normal Estimationcited by 122
  7. 20234M: Massively Multimodal Masked Modelingcited by 118
  8. 2025DriveSuprim: Towards Precise Trajectory Selection for End-to-End Planningcited by 91
  9. 2024Fine-Tuning Image-Conditional Diffusion Models is Easier than You Thinkcited by 82
  10. 2025Marigold: Affordable Adaptation of Diffusion-Based Image Generators for Image Analysiscited by 78
  11. 2024What Matters When Repurposing Diffusion Models for General Dense Perception Tasks?cited by 74
  12. 2024BRAVE: Broadening the visual encoding of vision-language modelscited by 68
38 more
  1. 2023Consistent123: One Image to Highly Consistent 3D Asset Using Case-Aware Diffusion Priorscited by 64
  2. 2025WorldMirror: Universal 3D World Reconstruction with Any-Prior Promptingcited by 60
  3. 20244M-21: An Any-to-Any Vision Model for Tens of Tasks and Modalitiescited by 52
  4. 2023RDumb: A simple approach that questions our progress in continual test-time adaptationcited by 50
  5. 2023Generative Models: What Do They Know? Do They Know Things? Let's Find Out!cited by 50
  6. 2023RoboDepth: Robust Out-of-Distribution Depth Estimation under Corruptionscited by 47
  7. 2022A Probabilistic Framework for Lifelong Test-Time Adaptationcited by 44
  8. 2023StyleGAN knows Normal, Depth, Albedo, and Morecited by 44
  9. 2024Is Synthetic Data all We Need? Benchmarking the Robustness of Models Trained with Synthetic Imagescited by 41
  10. 2023Exploiting Diffusion Prior for Generalizable Dense Predictioncited by 37
  11. 2025LBM: Latent Bridge Matching for Fast Image-to-Image Translationcited by 36
  12. 2024Zero-Shot Scene Reconstruction from Single Images with Deep Prior Assemblycited by 34
  13. 2026HY-World 2.0: A Multi-Modal World Model for Reconstructing, Generating, and Simulating 3D Worldscited by 34
  14. 2023Robustness of SAM: Segment Anything Under Corruptions and Beyondcited by 30
  15. 2024Video2Game: Real-time, Interactive, Realistic and Browser-Compatible Environment from a Single Videocited by 26
  16. 2024Is Your LiDAR Placement Optimized for 3D Scene Understanding?cited by 25
  17. 2025NormalCrafter: Learning Temporally Consistent Normals from Video Diffusion Priorscited by 25
  18. 2023GeoVLN: Learning Geometry-Enhanced Visual Representation with Slot Attention for Vision-and-Language Navigationcited by 24
  19. 2024The Entropy Enigma: Success and Failure of Entropy Minimizationcited by 24
  20. 2023JointNet: Extending Text-to-Image Diffusion for Dense Distribution Modelingcited by 23
  21. 2023UrbanIR: Large-Scale Urban Scene Inverse Rendering from a Single Videocited by 22
  22. 2025Relative Pose Estimation through Affine Corrections of Monocular Depth Priorscited by 18
  23. 2023Benchmarking Test-Time Adaptation against Distribution Shifts in Image Classificationcited by 17
  24. 2024Latent Intrinsics Emerge from Training to Relightcited by 17
  25. 2025Ranked Entropy Minimization for Continual Test-Time Adaptationcited by 17
  26. 2023Temporal Coherent Test-Time Optimization for Robust Video Classificationcited by 15
  27. 20233D Copy-Paste: Physically Plausible Object Insertion for Monocular 3D Detectioncited by 15
  28. 2025How Well Does GPT-4o Understand Vision? Evaluating Multimodal Foundation Models on Standard Computer Vision Taskscited by 14
  29. 2023Benchmarking Robustness in Neural Radiance Fieldscited by 13
  30. 2023Evaluation of Test-Time Adaptation Under Computational Time Constraintscited by 13
  31. 2023The RoboDepth Challenge: Methods and Advancements Towards Robust Depth Estimationcited by 13
  32. 2023Classification robustness to common optical aberrationscited by 13
  33. 2023RGB-based Category-level Object Pose Estimation via Decoupled Metric Scale Recoverycited by 13
  34. 2024Toward a Holistic Evaluation of Robustness in CLIP Modelscited by 13
  35. 2023HybridAugment++: Unified Frequency Spectra Perturbations for Model Robustnesscited by 12
  36. 2023DiG-IN: Diffusion Guidance for Investigating Networks -- Uncovering Classifier Differences Neuron Visualisations and Visual Counterfactual Explanationscited by 11
  37. 2024Dynamic Sparse Training versus Dense Training: The Unexpected Winner in Image Corruption Robustnesscited by 11
  38. 2024Buffer Anytime: Zero-Shot Video Depth and Normal from Image Priorscited by 11

50 most cited of 120 All 120 (JSON)