The repository impliments style transfer GAN architecture on videos using Transfer Learning on VGG-19 architecture.
To run the notebook, upload it to a kaggle envoirnment and add VGG-19 and dataset1, dataset2 in /input/ directory.
Demo video can be found here.
The following artistic image was applied on the video. A sample video frame is shown below:
Syed Muhammad Ammar Alam & Mehdi Raza Khorasani