mlsanityData to 7 Oct 2026

Benchmarking Robustness in Object Detection: Autonomous Driving when Winter is Coming

arXiv 1907.07484 ↗First posted 17 Jul 2019cs.CV cs.LG stat.MLcited by 332

Not on the map. Why?

Abstract (arXiv)

The ability to detect objects regardless of image distortions or weather conditions is crucial for real-world applications of deep learning like autonomous driving. We here provide an easy-to-use benchmark to assess how object detection models perform when image quality degrades. The three resulting benchmark datasets, termed Pascal-C, Coco-C and Cityscapes-C, contain a large variety of image corruptions. We show that a range of standard object detection models suffer a severe performance loss on corrupted images (down to 30--60% of the original performance). However, a simple data augmentation trick---stylizing the training images---leads to a substantial increase in robustness across corruption type, severity and dataset. We envision our comprehensive benchmark to track future progress towards building robust object detection models. Benchmark, code and data are publicly available.

Key facts. Cited by 332 papers in the map (data through 7 Oct 2026). First posted on arXiv 17 Jul 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 52; 2022: 44; 2023: 74; 2024: 59; 2025: 56; 2026: 47.

332 papers.

Sort:

By area:

  1. 2021SWAD: Domain Generalization by Seeking Flat Minimacited by 431
  2. 2023Q-Bench: A Benchmark for General-Purpose Foundation Models on Low-level Visioncited by 247
  3. 2022DeepFusion: Lidar-Camera Deep Fusion for Multi-Modal 3D Object Detectioncited by 224
  4. 2021Towards Causal Representation Learningcited by 207
  5. 2021Partial success in closing the gap between human and machine visioncited by 185
  6. 2023Q-Instruct: Improving Low-level Visual Abilities for Multi-modality Foundation Modelscited by 152
  7. 2022SHIFT: A Synthetic Driving Dataset for Continuous Multi-Task Domain Adaptationcited by 146
  8. 2022Domain Generalization by Mutual-Information Regularization with Pre-trained Modelscited by 133
  9. 2021The Norm Must Go On: Dynamic Unsupervised Domain Adaptation by Normalizationcited by 131
  10. 2023Robo3D: Towards Robust and Reliable 3D Perception against Corruptionscited by 128
  11. 20223D Common Corruptions and Data Augmentationcited by 120
  12. 2023Benchmarking Robustness of 3D Object Detection to Common Corruptions in Autonomous Drivingcited by 112
38 more
  1. 2022Physics-Informed Machine Learning: A Survey on Problems, Methods and Applicationscited by 88
  2. 2021OoD-Bench: Quantifying and Understanding Two Dimensions of Out-of-Distribution Generalizationcited by 86
  3. 2024Towards Open-ended Visual Quality Comparisoncited by 85
  4. 2023Stronger, Fewer, & Superior: Harnessing Vision Foundation Models for Domain Generalized Semantic Segmentationcited by 82
  5. 2021Generalize then Adapt: Source-Free Domain Adaptive Semantic Segmentationcited by 81
  6. 2022Benchmarking Robustness of 3D Point Cloud Recognition Against Common Corruptionscited by 78
  7. 2023Milestones in Autonomous Driving and Intelligent Vehicles: Survey of Surveyscited by 75
  8. 2024Wavelet Convolutions for Large Receptive Fieldscited by 74
  9. 2021Test-Time Adaptation to Distribution Shift by Confidence Maximization and Input Transformationcited by 67
  10. 2022Probable Domain Generalization via Quantile Risk Minimizationcited by 66
  11. 2021Lost in Pruning: The Effects of Pruning Neural Networks beyond Test Accuracycited by 62
  12. 2021On the Practicality of Deterministic Epistemic Uncertaintycited by 62
  13. 2021Visual Representation Learning Does Not Generalize Strongly Within the Same Domaincited by 60
  14. 2022Learning Domain Adaptive Object Detection with Probabilistic Teachercited by 57
  15. 2023Radar-Camera Fusion for Object Detection and Semantic Segmentation in Autonomous Driving: A Comprehensive Reviewcited by 57
  16. 2022An Efficient Domain-Incremental Learning Approach to Drive in All Weather Conditionscited by 54
  17. 2021Meaningfully Debugging Model Mistakes using Conceptual Counterfactual Explanationscited by 53
  18. 2022Assaying Out-Of-Distribution Generalization in Transfer Learningcited by 53
  19. 2021RobustNav: Towards Benchmarking Robustness in Embodied Navigationcited by 52
  20. 2021Source-Free Adaptation to Measurement Shift via Bottom-Up Feature Restorationcited by 50
  21. 2023C-SFDA: A Curriculum Learning Aided Self-Training Framework for Efficient Source Free Domain Adaptationcited by 49
  22. 2021A Systematic Review of Robustness in Deep Learning for Computer Vision: Mind the gap?cited by 48
  23. 2023RoboDepth: Robust Out-of-Distribution Depth Estimation under Corruptionscited by 47
  24. 2021OOD-CV: A Benchmark for Robustness to Out-of-Distribution Shifts of Individual Nuisances in Natural Imagescited by 46
  25. 2022ImageNet-X: Understanding Model Mistakes with Factor of Variation Annotationscited by 44
  26. 2021Robust and Accurate Object Detection via Adversarial Learningcited by 42
  27. 2021When Human Pose Estimation Meets Robustness: Adversarial Algorithms and Benchmarkscited by 41
  28. 2022Domain Adaptive Object Detection for Autonomous Driving under Foggy Weathercited by 41
  29. 2024Is Synthetic Data all We Need? Benchmarking the Robustness of Models Trained with Synthetic Imagescited by 41
  30. 2024An Intelligent Agentic System for Complex Image Restoration Problemscited by 41
  31. 2023Neither hype nor gloom do DNNs justicecited by 40
  32. 2022Attention Consistency on Visual Corruptions for Single-Source Domain Generalizationcited by 37
  33. 2023Understanding the Robustness of 3D Object Detection with Bird's-Eye-View Representations in Autonomous Drivingcited by 37
  34. 2021Benchmarking the Robustness of Spatial-Temporal Models Against Corruptionscited by 35
  35. 2023Auxiliary Losses for Learning Generalizable Concept-based Modelscited by 35
  36. 2021Improving robustness against common corruptions with frequency biased modelscited by 34
  37. 2023COCO-O: A Benchmark for Object Detectors under Natural Distribution Shiftscited by 34
  38. 2023Self-supervised Monocular Depth Estimation: Let's Talk About The Weathercited by 33

50 most cited of 332 All 332 (JSON)