|
gptkbp:instanceOf
|
convolutional neural network architecture family
|
|
gptkbp:activatedBy
|
gptkb:ReLU
|
|
gptkbp:author
|
gptkb:Andrew_Zisserman
gptkb:Karen_Simonyan
|
|
gptkbp:developedBy
|
gptkb:University_of_Oxford
gptkb:Visual_Geometry_Group
|
|
gptkbp:digestSize
|
224x224 RGB images
|
|
gptkbp:firstPublished
|
2014
|
|
https://www.w3.org/2000/01/rdf-schema#label
|
VGG family
|
|
gptkbp:includesModel
|
gptkb:VGG16
VGG19
|
|
gptkbp:influenced
|
gptkb:ResNet
gptkb:DenseNet
|
|
gptkbp:lastLayer
|
fully connected
|
|
gptkbp:level
|
16
19
|
|
gptkbp:notableFeature
|
max pooling layers
small 3x3 convolution filters
uniform architecture
|
|
gptkbp:notableFor
|
deep convolutional layers
|
|
gptkbp:notablePublication
|
gptkb:Very_Deep_Convolutional_Networks_for_Large-Scale_Image_Recognition
|
|
gptkbp:openSource
|
gptkb:TensorFlow
gptkb:Keras
gptkb:PyTorch
|
|
gptkbp:parameter
|
~138 million (VGG16)
|
|
gptkbp:usedFor
|
image classification
object detection
|
|
gptkbp:usedIn
|
ImageNet Large Scale Visual Recognition Challenge (ILSVRC) 2014
|
|
gptkbp:bfsParent
|
gptkb:VGG-13
|
|
gptkbp:bfsLayer
|
7
|