Have you ever wondered how smartphones camera filters work?
How do Snapchat or Instagram filters work?
Do you want to paint like Picasso?
If yes then you should know about The Neural Style Transfer Algorithm.
Style Transfer can be defined as the process of reconstruction of an image in the style of another image.
For detailed explanation please visit this blog
Python Tensorflow implementation of "A Neural Algorithm of Artistic Style," Gatys et al.
Python > 3.5
Numpy
PIL
Matplotlib
Tensorflow > 2.0
Keras
If you have dependency version issue use vnev
- Clone or download this repo.
- Make the vnev (recommended)
virtualenv mypython
- cd to the directory where requirements.txt is located
activate your virtualenv
mypthon\Scripts\activate
then
run: pip install -r requirements.txt in your shell
- open the
neural_style_transfer.ipynb
(if using Jupyter notebook) or
neural_style_transfer.py
file
change content_path
variable to path of your content image
style_path variable
to path of your style image - Run jupyter notebook or
python3 neural_style_transfer.py
in the shell.
Iterations - 1000 content_weight=1e-1,style_weight=1e2 Iterations - 500 content_weight=1e-1,style_weight=1e2
- Deploying NST model as a Django webapp.
visit this link for webapp
https://github.com/Shashi456/Neural-Style
https://www.tensorflow.org/tutorials/generative/style_transfer
https://github.com/bhrt-sharma/Neural-Style-Transfer