Week 5# Identification of Artists and Movements from Paintings with Machine Learning
This week, our main focus point is deep learning networks. For this reason, we have taken a look at the work that has been done in this field before. These include the neural networks, AlexNet, LeNet, Nejc Ilnec which won the kaggle competition, and the neural network developed by Jennie Chen and Andrew Deng.
Convolutional Neural Networks
A convolutional neural network (CNN) is a form of deep neural network used to analyze visual imagery in deep learning. It was essentially inspired by the biological processes. A CNN consists of an input layer, number of hidden layers and an output layer. Unlike conventional algorithms, which require hand-engineering of filters , CNN learns to optimize the filters through automatic learning.
LeNet-5
Yann Lecun created LeNet, one of the first convolutional neural networks, which drew attention to deep learning and helped to advance the field. This neural network, which was created in the early 1990s, has primarily been used for handwriting recognition.
There are 7 levels in LeNet-5: 3 convolutional layers, 2 subsampling layers, and 2 fully connected layers make up the layer composition.
AlexNet
AlexNet is a eight-layer convolutional neural network that can achieve high accuracies on even the most difficult datasets. It is developed by Alex Krizhevsky, Ilya Sutskever and Geoffrey Hinton. AlexNet took first place in the 2012 ImageNet competition, with a top-5 error rate of 15.3 percent, compared to 26.2 percent for second place.
LeNet5 & AlexNet Comparison
Nejc Ilenic
This convolutional neural network developed by Ilnec came first in the competition to determine whether the paintings in Kaggle were made by the same painter. The architecture developed by ilnec, which consists of many layers, can be found below.
Chen and Deng
This convolutional neural network, based on Ilnec’s model, was built on the recognition of which painter made the paintings. The architecture developed by Chen and Deng consists of 14 layers.
In addition to the above-mentioned works, our painting classification model will focus on both the relevant artist and movement .
Stay tuned…
