Attention for Image Registration (AiR): an unsupervised Transformer approach - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Preprints, Working Papers, ... Year :

Attention for Image Registration (AiR): an unsupervised Transformer approach

Abstract

Image registration as an important basis in signal processing task often encounter the problem of stability and efficiency. Non-learning registration approaches rely on the optimization of the similarity metrics between the fix- and moving images. Yet, those approaches are usually costly in both time and space complexity. The problem can be worse when the size of the image is large or the deformations between the images are severe. Recently, deep learning, or precisely saying, the convolutional neural network (CNN) based image registration methods have been widely investigated in the research community and show promising effectiveness to overcome the weakness of non-learning based methods. To explore the advanced learning approaches in image registration problem for solving practical issues, we present in this paper a method of introducing attention mechanism in deformable image registration problem. The proposed approach is based on learning the deformation field with a Transformer framework that does not rely on the CNN but can be efficiently trained on GPGPU devices also. Our method learns an artificially generated deformation map and be tested on a MINST dataset.
Fichier principal
Vignette du fichier
DRAFT__Attention_for_Image_Registration__A_Transformer (3).pdf (493.89 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03202230 , version 1 (19-04-2021)
hal-03202230 , version 2 (30-04-2021)
hal-03202230 , version 3 (06-05-2021)

Identifiers

  • HAL Id : hal-03202230 , version 3

Cite

Zihao Wang, Hervé Delingette. Attention for Image Registration (AiR): an unsupervised Transformer approach. 2021. ⟨hal-03202230v3⟩
205 View
496 Download

Share

Gmail Facebook Twitter LinkedIn More