EFANet: Exchangeable Feature Alignment Network for Arbitrary Style Transfer

Abstract

Style transfer has been an important topic both in computer vision and graphics. Since the seminal work of Gatys et al. first demonstrates the power of stylization through optimization in the deep feature space, quite a few approaches have achieved real-time arbitrary style transfer with straightforward statistic matching techniques. In this work, our key observation is that only considering features in the input style image for the global deep feature statistic matching or local patch swap may not always ensure a satisfactory style transfer; see e.g., Figure 1. Instead, we propose a novel transfer framework, EFANet, that aims to jointly analyze and better align exchangeable features extracted from content and style image pair. In this way, the style features from the style image seek for the best compatibility with the content information in the content image, leading to more structured stylization results. In addition, a new whitening loss is developed for purifying the computed content features and better fusion with styles in feature space. Qualitative and quantitative experiments demonstrate the advantages of our approach.

arXiv commentAccepted by AAAI 2020

References in corpus (19)

• A Learned Representation For Artistic Style
• Texture Networks: Feed-forward Synthesis of Textures and Stylized Images
• Domain Separation Networks
• DualGAN: Unsupervised Dual Learning for Image-to-Image Translation
• Multimodal Unsupervised Image-to-Image Translation
• Fast Patch-based Style Transfer of Arbitrary Style
• Stable and Controllable Neural Texture Synthesis and Style Transfer Using Histogram Losses
• DRIT++: Diverse Image-to-Image Translation via Disentangled Representations
• Neural Style Transfer: A Review
• Exemplar Guided Unsupervised Image-to-Image Translation with Semantic Consistency
• Multi-marginal Wasserstein GAN
• Adversarial Learning with Local Coordinate Coding
• Meta Networks for Neural Style Transfer
• DiDA: Disentangled Synthesis for Domain Adaptation
• Avatar-Net: Multi-scale Zero-shot Style Transfer by Feature Decoration
• A Style-Aware Content Loss for Real-time HD Style Transfer
• Controlling Perceptual Factors in Neural Style Transfer
• ETNet: Error Transition Network for Arbitrary Style Transfer
• Disentangled Representations in Neural Models