|
gptkbp:instanceOf
|
gptkb:convolutional_neural_network
|
|
gptkbp:activatedBy
|
gptkb:ReLU
|
|
gptkbp:architecture
|
encoder-decoder
|
|
gptkbp:author
|
gptkb:Thomas_Brox
gptkb:Olaf_Ronneberger
gptkb:Philipp_Fischer
|
|
gptkbp:citation
|
high
|
|
gptkbp:designedFor
|
biomedical image segmentation
|
|
gptkbp:finalActivationFunction
|
sigmoid
|
|
gptkbp:hasContractingPath
|
true
|
|
gptkbp:hasExpandingPath
|
true
|
|
gptkbp:hasSkipConnections
|
true
|
|
gptkbp:hasSymmetricStructure
|
true
|
|
gptkbp:hasVariant
|
gptkb:3D_U-Net
gptkb:Attention_U-Net
gptkb:Residual_U-Net
gptkb:U-Net++
|
|
https://www.w3.org/2000/01/rdf-schema#label
|
U-Net
|
|
gptkbp:influenced
|
medical image analysis
deep learning for segmentation
|
|
gptkbp:input
|
gptkb:illustrator
|
|
gptkbp:introduced
|
gptkb:Olaf_Ronneberger
|
|
gptkbp:introducedIn
|
2015
|
|
gptkbp:lossFunction
|
cross-entropy
dice loss
|
|
gptkbp:notableFeature
|
precise localization
skip connections between encoder and decoder
works with few training images
|
|
gptkbp:notablePublication
|
gptkb:U-Net:_Convolutional_Networks_for_Biomedical_Image_Segmentation
https://arxiv.org/abs/1505.04597
|
|
gptkbp:openSource
|
gptkb:TensorFlow
gptkb:Keras
gptkb:PyTorch
|
|
gptkbp:output
|
segmentation map
|
|
gptkbp:platform
|
deep learning
|
|
gptkbp:publishedIn
|
gptkb:MICCAI_2015
|
|
gptkbp:usedIn
|
remote sensing
autonomous driving
environmental monitoring
semantic segmentation
instance segmentation
agricultural image analysis
cell segmentation
histopathology image analysis
microscopy image analysis
organ segmentation
radiology image analysis
satellite image segmentation
tumor segmentation
|
|
gptkbp:bfsParent
|
gptkb:convolutional_neural_network
gptkb:Pix2Pix
|
|
gptkbp:bfsLayer
|
5
|