mlsanityData to 7 Oct 2026

Benchmarking Robustness of 3D Point Cloud Recognition Against Common Corruptions

arXiv 2201.12296 ↗First posted 28 Jan 2022cs.LG cs.AI cs.CVcited by 78

Not on the map. Why?

Abstract (arXiv)

Deep neural networks on 3D point cloud data have been widely used in the real world, especially in safety-critical applications. However, their robustness against corruptions is less studied. In this paper, we present ModelNet40-C, the first comprehensive benchmark on 3D point cloud corruption robustness, consisting of 15 common and realistic corruptions. Our evaluation shows a significant gap between the performances on ModelNet40 and ModelNet40-C for state-of-the-art (SOTA) models. To reduce the gap, we propose a simple but effective method by combining PointCutMix-R and TENT after evaluating a wide range of augmentation and test-time adaptation strategies. We identify a number of critical insights for future studies on corruption robustness in point cloud recognition. For instance, we unveil that Transformer-based architectures with proper training recipes achieve the strongest robustness. We hope our in-depth analysis will motivate the development of robust training strategies or architecture designs in the 3D point cloud domain. Our codebase and dataset are included in https://github.com/jiachens/ModelNet40-C

Key facts. Cited by 78 papers in the map (data through 7 Oct 2026). Builds on 52 of its 77 references that are in the map. First posted on arXiv 28 Jan 2022.

Builds on

52 of 77 references.

References · 52

  1. 2017Attention Is All You Needcited by 59,932
  2. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  3. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  4. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  5. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  6. 2015Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shiftcited by 6,979
  7. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,223
  8. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,384
  9. 2017ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenescited by 3,088
  10. 2015ShapeNet: An Information-Rich 3D Model Repositorycited by 2,880
  11. 2019Benchmarking Neural Network Robustness to Common Corruptions and Perturbationscited by 2,857
  12. 2018Dynamic Graph CNN for Learning on Point Cloudscited by 2,220
38 more
  1. 2020The Many Faces of Robustness: A Critical Analysis of Out-of-Distribution Generalizationcited by 1,842
  2. 2017Improved Regularization of Convolutional Neural Networks with Cutoutcited by 1,623
  3. 2020Point Transformercited by 1,462
  4. 2020Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attackscited by 1,442
  5. 2020WILDS: A Benchmark of in-the-Wild Distribution Shiftscited by 1,366
  6. 2019KPConv: Flexible and Deformable Convolution for Point Cloudscited by 1,202
  7. 2020Center-based 3D Object Detection and Trackingcited by 1,143
  8. 2018PointRCNN: 3D Object Proposal Generation and Detection from Point Cloudcited by 939
  9. 2019AugMix: A Simple Data Processing Method to Improve Robustness and Uncertaintycited by 897
  10. 2019PV-RCNN: Point-Voxel Feature Set Abstraction for 3D Object Detectioncited by 890
  11. 2018PointCNN: Convolution On X-Transformed Pointscited by 764
  12. 2020PCT: Point cloud transformercited by 730
  13. 2018AutoAugment: Learning Augmentation Policies from Datacited by 728
  14. 2018Open3D: A Modern Library for 3D Data Processingcited by 717
  15. 2020RobustBench: a standardized adversarial robustness benchmarkcited by 659
  16. 2018PointConv: Deep Convolutional Networks on 3D Point Cloudscited by 656
  17. 2019Revisiting Point Cloud Classification: A New Benchmark Dataset and Classification Model on Real-World Datacited by 532
  18. 20213D Shape Generation and Completion through Point-Voxel Diffusioncited by 516
  19. 2019On Evaluating Adversarial Robustnesscited by 442
  20. 2020Improving robustness against common corruptions by covariate shift adaptationcited by 427
  21. 2017Large-scale Point Cloud Semantic Segmentation with Superpoint Graphscited by 384
  22. 2019Relation-Shape Convolutional Neural Network for Point Cloud Analysiscited by 346
  23. 2019Benchmarking Robustness in Object Detection: Autonomous Driving when Winter is Comingcited by 332
  24. 2019Deep Closest Point: Learning Representations for Point Cloud Registrationcited by 318
  25. 2021Are Transformers More Robust Than CNNs?cited by 187
  26. 2021Revisiting Point Cloud Shape Classification with a Simple and Effective Baselinecited by 145
  27. 2018Generating 3D Adversarial Point Cloudscited by 144
  28. 2020PointMixup: Augmentation for Point Cloudscited by 104
  29. 2021On Interaction Between Augmentations and Corruptions in Natural Corruption Robustnesscited by 98
  30. 2020Towards Robust LiDAR-based Perception in Autonomous Driving: General Black-box Adversarial Sensor Attack and Countermeasurescited by 95
  31. 2019Extending Adversarial Attacks and Defenses to Deep 3D Point Cloud Classifierscited by 92
  32. 2018Point Cloud GANcited by 79
  33. 2021PointCutMix: Regularization Strategy for Point Cloud Classificationcited by 69
  34. 2021Regularization Strategy for Point Cloud via Rigidly Mixed Samplecited by 59
  35. 2021PointGuard: Provably Robust 3D Point Cloud Classificationcited by 49
  36. 2020On Isometry Robustness of Deep 3D Point Cloud Models under Adversarial Attackscited by 46
  37. 2021Defending Against Image Corruptions Through Adversarial Augmentationscited by 28
  38. 20213D Medical Point Transformer: Introducing Convolution to Attention Networks for Medical Point Cloud Analysiscited by 23

50 most cited of 52 All 52 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 16; 2023: 22; 2024: 14; 2025: 17; 2026: 9.

78 papers.

Sort:

By area:

  1. 2023Robo3D: Towards Robust and Reliable 3D Perception against Corruptionscited by 128
  2. 20223D Common Corruptions and Data Augmentationcited by 120
  3. 2023Benchmarking Robustness of 3D Object Detection to Common Corruptions in Autonomous Drivingcited by 112
  4. 2025DriveSuprim: Towards Precise Trajectory Selection for End-to-End Planningcited by 91
  5. 2022Benchmarking the Robustness of LiDAR-Camera Fusion for 3D Object Detectioncited by 49
  6. 2022Revisiting Realistic Test-Time Training: Sequential Inference and Adaptation by Anchored Clusteringcited by 44
  7. 2023OmniVec: Learning robust representations with cross modal sharingcited by 44
  8. 2023Understanding the Robustness of 3D Object Detection with Bird's-Eye-View Representations in Autonomous Drivingcited by 37
  9. 2023Continual Test-time Domain Adaptation via Dynamic Sample Selectioncited by 35
  10. 2023On the Robustness of Open-World Test-Time Training: Self-Training with Dynamic Prototype Expansioncited by 34
  11. 2023Towards Real-World Test-Time Adaptation: Tri-Net Self-Training with Balanced Normalizationcited by 34
  12. 2022Frame Mining: a Free Lunch for Learning Robotic Manipulation from 3D Point Cloudscited by 31
38 more
  1. 2023Robustness of SAM: Segment Anything Under Corruptions and Beyondcited by 30
  2. 2024GREAT: Geometry-Intention Collaborative Inference for Open-Vocabulary 3D Object Affordance Groundingcited by 26
  3. 2023Advancements in Point Cloud Data Augmentation for Deep Learning: A Surveycited by 24
  4. 2023Deep Learning-based 3D Point Cloud Classification: A Systematic Survey and Outlookcited by 24
  5. 2022MATE: Masked Autoencoders are Online 3D Test-Time Learnerscited by 23
  6. 2022PointDP: Diffusion-driven Purification against Adversarial Attacks on 3D Point Cloud Recognitioncited by 16
  7. 2023CALICO: Self-Supervised Camera-LiDAR Contrastive Pre-training for BEV Perceptioncited by 14
  8. 2024Backpropagation-free Network for 3D Test-time Adaptationcited by 14
  9. 2023Invariant Training 2D-3D Joint Hard Samples for Few-Shot Point Cloud Recognitioncited by 13
  10. 2025OmniVec2 -- A Novel Transformer based Network for Large Scale Multimodal and Multitask Learningcited by 13
  11. 20223DOS: Towards 3D Open Set Learning -- Benchmarking and Understanding Semantic Novelty Detection on Point Cloudscited by 11
  12. 2023VPA: Fully Test-Time Visual Prompt Adaptationcited by 11
  13. 2024CloudFixer: Test-Time Adaptation for 3D Point Clouds via Diffusion-Guided Geometric Transformationcited by 11
  14. 2022PointCAT: Contrastive Adversarial Training for Robust Point Cloud Recognitioncited by 10
  15. 2022Common Corruption Robustness of Point Cloud Detectors: Benchmark and Enhancementcited by 9
  16. 2022Point-DAE: Denoising Autoencoders for Self-supervised Point Cloud Learningcited by 9
  17. 2022PointCA: Evaluating the Robustness of 3D Point Cloud Completion Models Against Adversarial Examplescited by 9
  18. 2024Cocoon: Robust Multi-Modal Perception with Uncertainty-Aware Sensor Fusioncited by 9
  19. 2024Test-Time Adaptation in Point Clouds: Leveraging Sampling Variation with Weight Averagingcited by 7
  20. 2024Test-Time Adaptation of 3D Point Clouds via Denoising Diffusion Modelscited by 7
  21. 2023PointCert: Point Cloud Classification with Deterministic Certified Robustness Guaranteescited by 6
  22. 2023Risk-optimized Outlier Removal for Robust 3D Point Cloud Classificationcited by 6
  23. 2022On the Robustness of 3D Object Detectorscited by 5
  24. 2023Robust Point Cloud Processing through Positional Embeddingcited by 5
  25. 2023PointPatchMix: Point Cloud Mixing with Patch Scoringcited by 4
  26. 2023Tinto: Multisensor Benchmark for 3D Hyperspectral Point Cloud Segmentation in the Geosciencescited by 4
  27. 2023Neural Image Compression: Generalization, Robustness, and Spectral Biasescited by 4
  28. 2023Benchmarking and Analyzing Robust Point Cloud Recognition: Bag of Tricks for Defending Adversarial Examplescited by 4
  29. 2025Point-Cache: Test-time Dynamic and Hierarchical Cache for Robust and Generalizable Point Cloud Analysiscited by 4
  30. 2025SMART-PC: Skeletal Model Adaptation for Robust Test-Time Training in Point Cloudscited by 4
  31. 2023Benchmarking Robustness of AI-Enabled Multi-sensor Fusion Systems: Challenges and Opportunitiescited by 3
  32. 2024Advancements in Point Cloud-Based 3D Defect Detection and Classification for Industrial Systems: A Comprehensive Surveycited by 3
  33. 2024A Survey on Adversarial Robustness of LiDAR-based Machine Learning Perception in Autonomous Vehiclescited by 3
  34. 2025Purge-Gate: Backpropagation-Free Test-Time Adaptation for Point Clouds Classification via Token Purgingcited by 3
  35. 2024Assessing Visually-Continuous Corruption Robustness of Neural Networks Relative to Human Performancecited by 2
  36. 2024Exploring Human-in-the-Loop Test-Time Adaptation by Synergizing Active Learning and Model Selectioncited by 2
  37. 2024OODFace: Benchmarking Robustness of Face Recognition under Common Corruptions and Appearance Variationscited by 2
  38. 2024ObjVariantEnsemble: Advancing Point Cloud LLM Evaluation in Challenging Scenes with Subtly Distinguished Objectscited by 2

50 most cited of 78 All 78 (JSON)