GenHarris-ResNet: A Rotation Invariant Neural Network Based on Elementary Symmetric Polynomials - Morphologie mathématique (CMM) Access content directly
Conference Papers Year : 2023

GenHarris-ResNet: A Rotation Invariant Neural Network Based on Elementary Symmetric Polynomials

Valentin Penaud--Polge
  • Function : Author
  • PersonId : 1149801
Jesus Angulo

Abstract

In this paper, we propose a rotation invariant neural network based on Gaussian derivatives. The proposed network covers the main steps of the Harris corner detector in a generalized manner. More precisely, the Harris corner response function is a combination of the elementary symmetric polynomials of the integrated dyadic (outer) product of the gradient with itself. In the same way, we define matrices to be the self dyadic product of vectors composed with higher order partial derivatives and combine the elementary symmetric polynomials. A specific global pooling layer is used to mimic the local pooling used by Harris in his method. The proposed network is evaluated through three experiments. It first shows a quasi perfect invariance to rotations on Fashion-MNIST, it obtains competitive results compared to other rotation invariant networks on MNIST-Rot, and it obtains better performances classifying galaxies (EFIGI Dataset) than networks using up to a thousand times more trainable parameters.
Fichier principal
Vignette du fichier
SSVM_hal_archive-3.pdf (2.34 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-04117649 , version 1 (05-06-2023)

Identifiers

Cite

Valentin Penaud--Polge, Santiago Velasco-Forero, Jesus Angulo. GenHarris-ResNet: A Rotation Invariant Neural Network Based on Elementary Symmetric Polynomials. Scale Space and Variational Methods in Computer Vision, May 2023, Santa Margherita di Pula, Italy. pp.149-161, ⟨10.1007/978-3-031-31975-4_12⟩. ⟨hal-04117649⟩
12 View
40 Download

Altmetric

Share

Gmail Facebook X LinkedIn More