/Awesome-Style-Transfer-with-Deep-Learning

This repository presents style transfer resources that involves deep learning methods.

MIT LicenseMIT

Awesome-Style-Transfer-with-Deep-Learning Awesome

This repository presents style transfer resources that involves deep learning methods. T emophasis is placed on state-of-the-art papers for recent 2 years.

Papers

  • Recommended Paper is in Bold

2021

ICCV

  • Domain-Aware Universal Style Transfer [Paper] [Pytorch]

  • AdaAttN: Revisit Attention Mechanism in Arbitrary Neural Style Transfer [Paper] [Paddle] [Pytorch]

  • Manifold Alignment for Semantically Aligned Style Transfer [Paper] [Pytorch]

  • StyleFormer: Real-time Arbitrary Style Transfer via Parametric Style Composition [Paper]

  • Diverse Image Style Transfer via Invertible Cross-Space Mapping [Paper]

  • DRB-GAN: A Dynamic ResBlock Generative Adversarial Network for Artistic Style Transfer [Paper]

CVPR

  • ArtFlow: Unbiased Image Style Transfer via Reversible Neural Flows [Paper] [Pytorch]

  • Learning To Warp for Style Transfer [Paper] [Matlab]

  • Drafting and Revision: Laplacian Pyramid Network for Fast High-Quality Artistic Style Transfer [Paper]

  • DualAST: Dual Style-Learning Networks for Artistic Style Transfer [Paper] [Tensorflow]

  • In the Light of Feature Distributions: Moment Matching for Neural Style Transfer [Paper] [Pytorch]

  • Style-Aware Normalized Loss for Improving Arbitrary Style Transfer [Paper]

  • Rethinking Style Transfer: From Pixels to Parameterized Brushstrokes [Paper] [Tensorflow]

  • Adaptive Convolutions for Structure-Aware Style Transfer [Paper]

  • What Can Style Transfer and Paintings Do for Model Robustness? [Paper] [Pytorch]

  • Rethinking and Improving the Robustness of Image Style Transfer [Paper]

2020

Baseline Methods (Commonly Used)

Gatys (CVPR, 2016) A neural algorithm of artistic style

AdaIN (ICCV, 2017) Arbitrary style transfer in real-time with adaptive instance normalization

WCT (NIPS, 2017) Universal Style Transfer via Feature Transforms

LST (CVPR, 2019) Learning linear transformations for fast image and video style transfer

Avatar-Net (CVPR, 2018) Avatar-Net: Multi-scale Zero-shot Style Transfer by Feature Decoration

SANet (CVPR, 2019) Arbitrary Style Transfer with Style-Attentional Networks

STROTSS (CVPR, 2019) Style Transfer by Relaxed Optimal Transport and Self-Similarity

Surveys

Datasets

Metrics

Resources

Contact