|
gptkbp:instanceOf
|
deep learning model architecture
|
|
gptkbp:advantage
|
precise localization
works with few training images
high memory usage
limited to pixel-wise tasks
|
|
gptkbp:basedOn
|
gptkb:convolutional_neural_network
|
|
gptkbp:citation
|
over 50,000
|
|
gptkbp:hasComponent
|
skip connections
decoder
encoder
|
|
gptkbp:hasPaperTitle
|
gptkb:U-Net:_Convolutional_Networks_for_Biomedical_Image_Segmentation
|
|
gptkbp:hasVariant
|
gptkb:3D_U-Net
gptkb:Attention_U-Net
gptkb:Residual_U-Net
|
|
https://www.w3.org/2000/01/rdf-schema#label
|
U-Net architecture
|
|
gptkbp:input
|
gptkb:illustrator
|
|
gptkbp:inspiredBy
|
gptkb:fully_convolutional_networks
|
|
gptkbp:introduced
|
gptkb:Olaf_Ronneberger
|
|
gptkbp:introducedIn
|
2015
|
|
gptkbp:notableFor
|
biomedical image segmentation
|
|
gptkbp:openSource
|
gptkb:TensorFlow
gptkb:Keras
gptkb:PyTorch
|
|
gptkbp:output
|
segmentation map
|
|
gptkbp:publishedIn
|
gptkb:MICCAI_2015
|
|
gptkbp:shape
|
U-shaped
|
|
gptkbp:usedFor
|
image segmentation
|
|
gptkbp:usedIn
|
medical imaging
cell segmentation
satellite image analysis
|
|
gptkbp:bfsParent
|
gptkb:U-Net:_Convolutional_Networks_for_Biomedical_Image_Segmentation
|
|
gptkbp:bfsLayer
|
7
|