Pixel-wise Smoothing for Certified Robustness against Camera Motion Perturbations
Not on the map. Why?
Abstract (arXiv)
Deep learning-based visual perception models lack robustness when faced with camera motion perturbations in practice. The current certification process for assessing robustness is costly and time-consuming due to the extensive number of image projections required for Monte Carlo sampling in the 3D camera motion space. To address these challenges, we present a novel, efficient, and practical framework for certifying the robustness of 3D-2D projective transformations against camera motion perturbations. Our approach leverages a smoothing distribution over the 2D pixel space instead of in the 3D physical space, eliminating the need for costly camera motion sampling and significantly enhancing the efficiency of robustness certifications. With the pixel-wise smoothed classifier, we are able to fully upper bound the projection errors using a technique of uniform partitioning in camera motion space. Additionally, we extend our certification framework to a more general scenario where only a single-frame point cloud is required in the projection oracle. Through extensive experimentation, we validate the trade-off between effectiveness and efficiency enabled by our proposed method. Remarkably, our approach achieves approximately 80% certified accuracy while utilizing only 30% of the projected image frames. The code is available at https://github.com/HanjiangHu/pixel-wise-smoothing.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 54 of its 76 references that are in the map. First posted on arXiv 22 Sep 2023.
Builds on
54 of 76 references.
References · 54
- 2015Deep Residual Learning for Image Recognitioncited by 51,809
- 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,185
- 2014Explaining and Harnessing Adversarial Examplescited by 7,287
- 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
- 2013Intriguing properties of neural networkscited by 5,114
- 2019Benchmarking Neural Network Robustness to Common Corruptions and Perturbationscited by 2,857
- 2019Certified Adversarial Robustness via Randomized Smoothingcited by 1,324
- 2014Depth Map Prediction from a Single Image using a Multi-Scale Deep Networkcited by 1,264
- 2017Ensemble Adversarial Training: Attacks and Defensescited by 833
- 2017Reluplex: An Efficient SMT Solver for Verifying Deep Neural Networkscited by 739
- 2017Provable defenses against adversarial examples via the convex outer adversarial polytopecited by 564
- 2020On Adaptive Attacks to Adversarial Example Defensescited by 547
38 more
- 2018Efficient Neural Network Robustness Certification with General Activation Functionscited by 439
- 2018Defense-GAN: Protecting Classifiers Against Adversarial Attacks Using Generative Modelscited by 362
- 2019Provably Robust Deep Learning via Adversarially Trained Smoothed Classifierscited by 328
- 2018Certified Defenses against Adversarial Examplescited by 305
- 2018Characterizing Adversarial Subspaces Using Local Intrinsic Dimensionalitycited by 280
- 2018Generating Adversarial Examples with Adversarial Networkscited by 256
- 2017Formal Verification of Piece-Wise Linear Feed-Forward Neural Networkscited by 241
- 2019Algorithms for Verifying Deep Neural Networkscited by 221
- 2018Semidefinite relaxations for certifying robustness to adversarial examplescited by 178
- 2022SHIFT: A Synthetic Driving Dataset for Continuous Multi-Task Domain Adaptationcited by 146
- 2022(Certified!!) Adversarial Robustness for Free!cited by 130
- 2019A Convex Relaxation Barrier to Tight Robustness Verification of Neural Networkscited by 124
- 2022Robustness and Accuracy Could Be Reconcilable by (Proper) Definitioncited by 109
- 2022General Cutting Planes for Bound-Propagation-Based Neural Network Verificationcited by 102
- 2020SoK: Certified Robustness for Deep Neural Networkscited by 100
- 2021On Interaction Between Augmentations and Corruptions in Natural Corruption Robustnesscited by 98
- 2018Semantic Adversarial Examplescited by 86
- 2021PRIMA: General and Precise Neural Network Certification via Scalable Convex Hull Approximationscited by 77
- 2020Consistency Regularization for Certified Robustness of Smoothed Classifierscited by 73
- 2020Enabling certification of verification-agnostic networks via memory-efficient semidefinite programmingcited by 65
- 2022SafeBench: A Benchmarking Platform for Safety Evaluation of Autonomous Vehiclescited by 65
- 2020Certified Defense to Image Transformations via Randomized Smoothingcited by 51
- 2020TSS: Transformation-Specific Smoothing for Robustness Certificationcited by 49
- 2017Geometric robustness of deep networks: analysis and improvementcited by 45
- 2020DASGIL: Domain Adaptation for Semantic and Geometric-aware Image-based Localizationcited by 28
- 2022GSmooth: Certified Robustness against Semantic Transformations via Generalized Randomized Smoothingcited by 23
- 2021Investigating the Impact of Multi-LiDAR Placement on Object Detection for Autonomous Drivingcited by 22
- 2017Towards Practical Verification of Machine Learning: The Case of Computer Vision Systemscited by 19
- 2020Improved Handling of Motion Blur in Online Object Detectioncited by 18
- 2021DeformRS: Certifying Input Deformations with Randomized Smoothingcited by 17
- 2021Robustness Certification for Point Cloud Modelscited by 16
- 2018Semantic Adversarial Deep Learningcited by 15
- 2022Provable Defense Against Geometric Transformationscited by 14
- 2023The RoboDepth Challenge: Methods and Advancements Towards Robust Depth Estimationcited by 13
- 2018Where Should We Place LiDARs on the Autonomous Vehicle? - An Optimal Design Approachcited by 12
- 2020Efficient Certification of Spatial Robustnesscited by 11
- 2022TPC: Transformation-Specific Smoothing for Point Cloud Modelscited by 10
- 2022CARE: Certifiably Robust Learning with Reasoning via Variational Inferencecited by 10
50 most cited of 54 All 54 (JSON)
Built on by
Citing papers per year: 2025: 1; 2026: 1.
2 papers.