DeformRS: Certifying Input Deformations with Randomized Smoothing

• URL: http://arxiv.org/abs/2107.00996v1
• Date: Fri, 2 Jul 2021 12:20:15 GMT
• Title: DeformRS: Certifying Input Deformations with Randomized Smoothing
• Authors: Motasem Alfarra, Adel Bibi, Naeemullah Khan, Philip H. S. Torr, and Bernard Ghanem
• Abstract summary: 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.
• Score: 121.88209420825582
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Deep neural networks are vulnerable to input deformations in the form of vector fields of pixel displacements and to other parameterized geometric deformations e.g. translations, rotations, etc. Current input deformation certification methods either (i) do not scale to deep networks on large input datasets, or (ii) can only certify a specific class of deformations, e.g. only rotations. We reformulate certification in randomized smoothing setting for both general vector field and parameterized deformations and propose DeformRS-VF and DeformRS-Par, respectively. Our new formulation scales to large networks on large input datasets. For instance, DeformRS-Par certifies rich deformations, covering translations, rotations, scaling, affine deformations, and other visually aligned deformations such as ones parameterized by Discrete-Cosine-Transform basis. Extensive experiments on MNIST, CIFAR10 and ImageNet show that DeformRS-Par outperforms existing state-of-the-art in certified accuracy, e.g. improved certified accuracy of 6% against perturbed rotations in the set [-10,10] degrees on ImageNet.

Related papers

• 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)
• RecRecNet: Rectangling Rectified Wide-Angle Images by Thin-Plate Spline Model and DoF-based Curriculum Learning [62.86400614141706]
  We propose a new learning model, i.e., Rectangling Rectification Network (RecRecNet) Our model can flexibly warp the source structure to the target domain and achieves an end-to-end unsupervised deformation. Experiments show the superiority of our solution over the compared methods on both quantitative and qualitative evaluations.
  arXiv  Detail & Related papers  (2023-01-04T15:12:57Z)
• 3DeformRS: Certifying Spatial Deformations on Point Clouds [61.62846778591536]
  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.
  arXiv  Detail & Related papers  (2022-04-12T10:24:31Z)
• Extracting Deformation-Aware Local Features by Learning to Deform [3.364554138758565]
  We present a new approach to compute features from still images that are robust to non-rigid deformations. We train the model architecture end-to-end by applying non-rigid deformations to objects in a simulated environment. Experiments show that our method outperforms state-of-the-art handcrafted, learning-based image, and RGB-D descriptors in different datasets.
  arXiv  Detail & Related papers  (2021-11-20T15:46:33Z)
• A Learning Framework for Diffeomorphic Image Registration based on Quasi-conformal Geometry [1.2891210250935146]
  We propose the quasi-conformal registration network (QCRegNet), an unsupervised learning framework, to obtain diffeomorphic 2D image registrations. QCRegNet consists of the estimator network and the Beltrami solver network (BSNet) Results show that the registration accuracy is comparable to state-of-the-art methods and diffeomorphism is to a great extent guaranteed.
  arXiv  Detail & Related papers  (2021-10-20T14:23:24Z)
• Augmenting Implicit Neural Shape Representations with Explicit Deformation Fields [95.39603371087921]
  Implicit neural representation is a recent approach to learn shape collections as zero level-sets of neural networks. We advocate deformation-aware regularization for implicit neural representations, aiming at producing plausible deformations as latent code changes.
  arXiv  Detail & Related papers  (2021-08-19T22:07:08Z)
• Learnable Gabor modulated complex-valued networks for orientation robustness [4.024850952459758]
  Learnable Gabor Convolutional Networks (LGCNs) are parameter-efficient and offer increased model complexity. We investigate the robustness of complex valued convolutional weights with learned Gabor filters to enable orientation transformations.
  arXiv  Detail & Related papers  (2020-11-23T21:22:27Z)
• Learning to Learn Parameterized Classification Networks for Scalable Input Images [76.44375136492827]
  Convolutional Neural Networks (CNNs) do not have a predictable recognition behavior with respect to the input resolution change. We employ meta learners to generate convolutional weights of main networks for various input scales. We further utilize knowledge distillation on the fly over model predictions based on different input resolutions.
  arXiv  Detail & Related papers  (2020-07-13T04:27:25Z)

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.