Style transfer

Computer vision

Style transfer is the process of transferring some visual features from one image to another image while preserving the latter’s content information. Since both these notions may be considered subjective, the problem of style transfer is not well defined and may be approached in many ways.

Style transfer with CNNs

This is an early example of style transfer with convolutional neural networks: (Gatys et al. 2016)


  1. . . "Image Style Transfer Using Convolutional Neural Networks". In 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2414–23. Las Vegas, NV, USA: IEEE. DOI.


← Back to Notes