recognizing styles of painitngs of different artists through residual neural networks
RESNET: The ResNet has input dimensions of 224×224×3.Its architecture is composed of different blocks, where each block uses a ”shortcut connection”. This shortcut connection can be a simple identity connection (id-block), or a connection with a convolutional layer (conv-block). The shortcut-ed part of the block uses 3 convolutions, with various number of filters for each block. The first and third convolution of this group use filters sizes of 1 x 1, and the second one is usually 3 x 3. At the end of a block, the features of the shorcut part and the shortcut-ed part are added. Each convolutional layer is followed by a batch-normalization layer.
The ResNet50 starts with a convolutional layer with a filter size of 7x7, generating 64 filters, followed by a batch-normalization layer, an activation layer and a max-pooling layer. Then there are 4 groups of blocks, each starting with a conv-block(3 convolutional layers). Each group contains respectively 1, 3, 5 and 2 id-blocks. The ResNet50 contains in total 53 (3 +(1+3+5+2)) convolutional layers.