equivariant_image_recon
None
view repo
In recent years the use of convolutional layers to encode an inductive bias (translational equivariance) in neural networks has proven to be a very fruitful idea. The successes of this approach have motivated a line of research into incorporating other symmetries into deep learning methods, in the form of group equivariant convolutional neural networks. Much of this work has been focused on roto-translational symmetry of π^d, but other examples are the scaling symmetry of π^d and rotational symmetry of the sphere. In this work, we demonstrate that group equivariant convolutional operations can naturally be incorporated into learned reconstruction methods for inverse problems that are motivated by the variational regularisation approach. Indeed, if the regularisation functional is invariant under a group symmetry, the corresponding proximal operator will satisfy an equivariance property with respect to the same group symmetry. As a result of this observation, we design learned iterative methods in which the proximal operators are modelled as group equivariant convolutional neural networks. We use roto-translationally equivariant operations in the proposed methodology and apply it to the problems of low-dose computerised tomography reconstruction and subsampled magnetic resonance imaging reconstruction. The proposed methodology is demonstrated to improve the reconstruction quality of a learned reconstruction method with a little extra computational cost at training time but without any extra cost at test time.
READ FULL TEXT
Model-based learned iterative reconstruction methods have recently been ...
read it
Data assisted reconstruction algorithms, incorporating trained neural
ne...
read it
Deep convolutional neural networks trained on large datsets have emerged...
read it
We survey the mathematical foundations of geometric deep learning, focus...
read it
Deep neural networks have been applied successfully to a wide variety of...
read it
Training machine learning models that can learn complex spatiotemporal
d...
read it
The robust and automated determination of crystal symmetry is of utmost
...
read it
None
Comments
There are no comments yet.