Pré-Publication, Document De Travail Année : 2022

On the Benefits of SO(3)-Equivariant Neural Networks for Spherical Image Processing

Simon Martin
  • Fonction : Auteur
  • PersonId : 1159112

Résumé

This paper deals with the use of SO(3)-Equivariant Neural Networks for processing spherical images such as those acquired with bi-directional fish-eye lenses. By using Generative Adversarial Networks trained on the Planesnet dataset to generate augmented training sets, we first confirm recent results according to which the use of SO(3)-equivariance mechanisms is more efficient than training data augmentation techniques for processing data with a native spherical geometry. We then explain how the action of SO(3) on the 2d-sphere can be projected on tangent planes, leading to an equivalent reformulation of the SO(3)-convolution operator in euclidean spaces and allowing to achieve a wider range of robustness properties through the design of appropriate local pooling mechanisms. Finally, we articulate our projected SO(3) convolution with recent works on fish-eye image processing in which SU(1, 1) and SL(2, R) Equivariant Neural Networks are coupled with hyperbolic projection mechanisms, drawing the path toward a consolidated approach for processing inputs represented as signals on homogeneous spaces.
Fichier principal
Vignette du fichier
spherical_equiv_benefit_martinlagrave.pdf (1) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03763121 , version 1 (29-08-2022)

Identifiants

  • HAL Id : hal-03763121 , version 1

Citer

Simon Martin, Pierre-Yves Lagrave. On the Benefits of SO(3)-Equivariant Neural Networks for Spherical Image Processing. 2022. ⟨hal-03763121⟩
128 Consultations
130 Téléchargements

Partager

More