mlsanityData to 7 Oct 2026

H-NeXt: The next step towards roto-translation invariant networks

arXiv 2311.01111 ↗First posted 2 Nov 2023cs.CV cs.LGcited by 1

Not on the map. Why?

Abstract (arXiv)

The widespread popularity of equivariant networks underscores the significance of parameter efficient models and effective use of training data. At a time when robustness to unseen deformations is becoming increasingly important, we present H-NeXt, which bridges the gap between equivariance and invariance. H-NeXt is a parameter-efficient roto-translation invariant network that is trained without a single augmented image in the training set. Our network comprises three components: an equivariant backbone for learning roto-translation independent features, an invariant pooling layer for discarding roto-translation information, and a classification layer. H-NeXt outperforms the state of the art in classification on unaugmented training sets and augmented test sets of MNIST and CIFAR-10.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 29 of its 51 references that are in the map. First posted on arXiv 2 Nov 2023.

Builds on

29 of 51 references.

References · 29

  1. 2022A ConvNet for the 2020scited by 3,545
  2. 2013Visualizing and Understanding Convolutional Networkscited by 2,278
  3. 2015Spatial Transformer Networkscited by 1,423
  4. 2016Group Equivariant Convolutional Networkscited by 1,154
  5. 2023ConvNeXt V2: Co-designing and Scaling ConvNets with Masked Autoencoderscited by 703
  6. 2018Spherical CNNscited by 452
  7. 2019General E(2)-Equivariant Steerable CNNscited by 441
  8. 2021Vector Neurons: A General Framework for SO(3)-Equivariant Networkscited by 304
  9. 2016Steerable CNNscited by 298
  10. 2018On the Generalization of Equivariance and Convolution in Neural Networks to the Action of Compact Groupscited by 296
  11. 2016Harmonic Networks: Deep Translation and Rotation Equivariancecited by 293
  12. 2019Gauge Equivariant Convolutional Networks and the Icosahedral CNNcited by 262
17 more
  1. 2014Understanding image representations by measuring their equivariance and equivalencecited by 240
  2. 2021Rethinking and Improving Relative Position Encoding for Vision Transformercited by 206
  3. 2021ReDet: A Rotation-equivariant Detector for Aerial Object Detectioncited by 202
  4. 2017Learning Steerable Filters for Rotation Equivariant CNNscited by 193
  5. 2016Rotation equivariant vector field networkscited by 98
  6. 2016Exploiting Cyclic Symmetry in Convolutional Neural Networkscited by 91
  7. 2018Roto-Translation Covariant Convolutional Networks for Medical Image Analysiscited by 91
  8. 2017Polar Transformer Networkscited by 81
  9. 2014Rigid-Motion Scattering for Texture Classificationcited by 52
  10. 2014Deep Roto-Translation Scattering for Object Classificationcited by 49
  11. 2020PDO-eConvs: Partial Differential Operator Based Equivariant Convolutionscited by 34
  12. 2018HexaConvcited by 27
  13. 2016Learning rotation invariant convolutional filters for texture classificationcited by 24
  14. 2021Exploiting Redundancy: Separable Group Convolutional Networks on Lie Groupscited by 22
  15. 2017Graph-based Isometry Invariant Representation Learningcited by 7
  16. 2022Learning Invariant Representations for Equivariant Neural Networks Using Orthogonal Momentscited by 3
  17. 2021Equivariance-bridged SO(2)-Invariant Representation Learning using Graph Convolutional Networkcited by 2

Built on by

Cited by year

Citing papers per year: 2024: 1.

1 paper.

  1. 2024Harmformer: Harmonic Networks Meet Transformers for Continuous Roto-Translation Equivariancecited by 1