3DeformRS: Certifying Spatial Deformations on Point Clouds

• URL: http://arxiv.org/abs/2204.05687v1
• Date: Tue, 12 Apr 2022 10:24:31 GMT
• Title: 3DeformRS: Certifying Spatial Deformations on Point Clouds
• Authors: Gabriel P\'erez S., Juan C. P\'erez, Motasem Alfarra, Silvio Giancola, Bernard Ghanem
• Abstract summary: 3D computer vision models are commonly used in security-critical applications such as autonomous driving and surgical robotics. Emerging concerns over the robustness of these models against real-world deformations must be addressed practically and reliably. We propose 3DeformRS, a method to certify the robustness of point cloud Deep Neural Networks (DNNs) against real-world deformations.
• Score: 61.62846778591536
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: 3D computer vision models are commonly used in security-critical applications such as autonomous driving and surgical robotics. Emerging concerns over the robustness of these models against real-world deformations must be addressed practically and reliably. In this work, we propose 3DeformRS, a method to certify the robustness of point cloud Deep Neural Networks (DNNs) against real-world deformations. We developed 3DeformRS by building upon recent work that generalized Randomized Smoothing (RS) from pixel-intensity perturbations to vector-field deformations. In particular, we specialized RS to certify DNNs against parameterized deformations (e.g. rotation, twisting), while enjoying practical computational costs. We leverage the virtues of 3DeformRS to conduct a comprehensive empirical study on the certified robustness of four representative point cloud DNNs on two datasets and against seven different deformations. Compared to previous approaches for certifying point cloud DNNs, 3DeformRS is fast, scales well with point cloud size, and provides comparable-to-better certificates. For instance, when certifying a plain PointNet against a 3{\deg} z-rotation on 1024-point clouds, 3DeformRS grants a certificate 3x larger and 20x faster than previous work.

Related papers

• Adaptive Point Transformer [88.28498667506165]
  Adaptive Point Cloud Transformer (AdaPT) is a standard PT model augmented by an adaptive token selection mechanism. AdaPT dynamically reduces the number of tokens during inference, enabling efficient processing of large point clouds.
  arXiv  Detail & Related papers  (2024-01-26T13:24:45Z)
• Learning-Based Biharmonic Augmentation for Point Cloud Classification [79.13962913099378]
  Biharmonic Augmentation (BA) is a novel and efficient data augmentation technique. BA diversifies point cloud data by imposing smooth non-rigid deformations on existing 3D structures. We present AdvTune, an advanced online augmentation system that integrates adversarial training.
  arXiv  Detail & Related papers  (2023-11-10T14:04:49Z)
• Sample-adaptive Augmentation for Point Cloud Recognition Against Real-world Corruptions [20.95456179904285]
  We propose an alternative to make sample-adaptive transformations based on the structure of the sample, named as AdaptPoint. A discriminator is utilized to prevent the generation of excessive corruption that deviates from the original data distribution. Experiments show that our method achieves state-of-the-art results on multiple corruption benchmarks, including ModelNet-C, our ScanObjectNN-C, and ShapeNet-C.
  arXiv  Detail & Related papers  (2023-09-19T08:46:43Z)
• SVNet: Where SO(3) Equivariance Meets Binarization on Point Cloud Representation [65.4396959244269]
  The paper tackles the challenge by designing a general framework to construct 3D learning architectures. The proposed approach can be applied to general backbones like PointNet and DGCNN. Experiments on ModelNet40, ShapeNet, and the real-world dataset ScanObjectNN, demonstrated that the method achieves a great trade-off between efficiency, rotation, and accuracy.
  arXiv  Detail & Related papers  (2022-09-13T12:12:19Z)
• 3DVerifier: Efficient Robustness Verification for 3D Point Cloud Models [17.487852393066458]
  Existing verification method for point cloud model is time-expensive and computationally unattainable on large networks. We propose 3DVerifier to tackle both challenges by adopting a linear relaxation function to bound the multiplication layer and combining forward and backward propagation. Our approach achieves an orders-of-magnitude improvement in verification efficiency for the large network, and the obtained certified bounds are also significantly tighter than the state-of-the-art verifiers.
  arXiv  Detail & Related papers  (2022-07-15T15:31:16Z)
• TPC: Transformation-Specific Smoothing for Point Cloud Models [9.289813586197882]
  We propose a transformation-specific smoothing framework TPC, which provides robustness guarantees for point cloud models against semantic transformation attacks. Experiments on several common 3D transformations show that TPC significantly outperforms the state of the art.
  arXiv  Detail & Related papers  (2022-01-30T05:41:50Z)
• DeformRS: Certifying Input Deformations with Randomized Smoothing [121.88209420825582]
  Deep neural networks are vulnerable to input deformations in the form of vector fields of pixel displacements. We reformulate certification in randomized smoothing setting for both general vector field and parameterized deformations. Our new formulation scales to large networks on large input datasets.
  arXiv  Detail & Related papers  (2021-07-02T12:20:15Z)
• Robustness Certification for Point Cloud Models [10.843109238068982]
  We introduce 3DCertify, the first verifier able to certify robustness of point cloud models. 3DCertify is based on two key insights: (i) a generic relaxation based on first-order Taylor approximations, and (ii) a precise relaxation for global feature pooling. We demonstrate the effectiveness of 3DCertify by performing an extensive evaluation on a wide range of 3D transformations.
  arXiv  Detail & Related papers  (2021-03-30T19:52:07Z)
• Adjoint Rigid Transform Network: Task-conditioned Alignment of 3D Shapes [86.2129580231191]
  Adjoint Rigid Transform (ART) Network is a neural module which can be integrated with a variety of 3D networks. ART learns to rotate input shapes to a learned canonical orientation, which is crucial for a lot of tasks. We will release our code and pre-trained models for further research.
  arXiv  Detail & Related papers  (2021-02-01T20:58:45Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.