Researchers from the Oxford Visual Geometry Group, or the VGG for short, developed the VGG network, which is characterized by its simplicity, using only 3 x 3 convolutional layers stacked on top of each other in increasing depth. Reducing volume size is handled by max pooling. At the end, two fully connected layers, each with 4,096 nodes, are then followed by a softmax layer. The only preprocessing done to the input is the subtraction of the mean RGB value, computed on the training set, from each pixel.

Pooling is carried out by max pooling layers, which follow some of the convolution layers. Not all the convolution layers are followed by max pooling. Max pooling is performed over a 2 x 2 pixel window, with a stride of 2. ReLU activation is used in each of the hidden layers. The number of filters increases with depth in most VGG variants. The 16-layered architecture VGG-16 is shown in the following diagram. The 19-layered architecture with uniform 3 x 3 convolutions (VGG-19) is shown along with ResNet in the following section. The success of VGG models confirms the importance of depth in image representations: