Caffe implementation of "Perceptual Losses for Real-Time Style Transfer and Super-Resolution"

Fast artistic style transfer by using feed forward network.

Requirement

• Caffe

Convert

python caffe_style_transfer.py <input_image_path> -m <model_path> -o <output_image_path> -g <use_gpu ? gpu_id : 0>

This repo has pretrained models as an example in model_caffe.

• example:

python caffe_style_transfer.py sample_images/HKU.jpg -m model_caffe/japan.caffemodel -o sample_images/output.jpg

or

python caffe_style_transfer.py sample_images/HKU.jpg -m model_caffe/modern.caffemodel -o sample_images/output.jpg