Neural Artistic Style in Python

Implementation of A Neural Algorithm of Artistic Style. A method to transfer the style of one image to the subject of another image.

Requirements

• DeepPy, Deep learning in Python.
• CUDArray with cuDNN, CUDA-accelerated NumPy.
• Pretrained VGG 19 model, choose imagenet-vgg-verydeep-19.

Examples

Execute

python neural_artistic_style.py --subject images/tuebingen.jpg --style images/starry_night.jpg

The two inputs are

Subject: Style:

The output becomes:

We can also choose a (younger version) of HM the Queen of Denmark as subject and paint her using different styles. Click the images to see the full size.

Subject

Styles

Outputs

Help

List command line options with

python neural_artistic_style.py --help