Scale Equivariant U-Net

by   Mateus Sangalli, et al.

In neural networks, the property of being equivariant to transformations improves generalization when the corresponding symmetry is present in the data. In particular, scale-equivariant networks are suited to computer vision tasks where the same classes of objects appear at different scales, like in most semantic segmentation tasks. Recently, convolutional layers equivariant to a semigroup of scalings and translations have been proposed. However, the equivariance of subsampling and upsampling has never been explicitly studied even though they are necessary building blocks in some segmentation architectures. The U-Net is a representative example of such architectures, which includes the basic elements used for state-of-the-art semantic segmentation. Therefore, this paper introduces the Scale Equivariant U-Net (SEU-Net), a U-Net that is made approximately equivariant to a semigroup of scales and translations through careful application of subsampling and upsampling layers and the use of aforementioned scale-equivariant layers. Moreover, a scale-dropout is proposed in order to improve generalization to different scales in approximately scale-equivariant architectures. The proposed SEU-Net is trained for semantic segmentation of the Oxford Pet IIIT and the DIC-C2DH-HeLa dataset for cell segmentation. The generalization metric to unseen scales is dramatically improved in comparison to the U-Net, even when the U-Net is trained with scale jittering, and to a scale-equivariant architecture that does not perform upsampling operators inside the equivariant pipeline. The scale-dropout induces better generalization on the scale-equivariant models in the Pet experiment, but not on the cell segmentation experiment.


page 2

page 5

page 7

page 8

page 13

page 14


Multi-Resolution Networks for Semantic Segmentation in Whole Slide Images

Digital pathology provides an excellent opportunity for applying fully c...

Lightweight and Progressively-Scalable Networks for Semantic Segmentation

Multi-scale learning frameworks have been regarded as a capable class of...

MSRF-Net: A Multi-Scale Residual Fusion Network for Biomedical Image Segmentation

Methods based on convolutional neural networks have improved the perform...

Autofocus Layer for Semantic Segmentation

We propose the autofocus convolutional layer for semantic segmentation w...

EVC-Net: Multi-scale V-Net with Conditional Random Fields for Brain Extraction

Brain extraction is one of the first steps of pre-processing 3D brain MR...

Atrous Space Bender U-Net (ASBU-Net/LogiNet)

With recent advances in CNNs, exceptional improvements have been made in...

Crack Semantic Segmentation using the U-Net with Full Attention Strategy

Structures suffer from the emergence of cracks, therefore, crack detecti...

Please sign up or login with your details

Forgot password? Click here to reset