Deep Photo Style Transfer
arXiv:1703.07511
Abstract
This paper introduces a deep-learning approach to photographic style transfer that handles a large variety of image content while faithfully transferring the reference style. Our approach builds upon the recent work on painterly transfer that separates style from the content of an image by considering different layers of a neural network. However, as is, this approach is not suitable for photorealistic style transfer. Even when both the input and reference images are photographs, the output still exhibits distortions reminiscent of a painting. Our contribution is to constrain the transformation from the input to the output to be locally affine in colorspace, and to express this constraint as a custom fully differentiable energy term. We show that this approach successfully suppresses distortion and yields satisfying photorealistic style transfers in a broad variety of scenarios, including transfer of the time of day, weather, season, and artistic edits.
References in corpus (1)
Cited by in corpus (41)
- Learning Linear Transformations for Fast Arbitrary Style Transfer
- GAN-based Virtual Re-Staining: A Promising Solution for Whole Slide Image Analysis
- Audio style transfer
- A Closed-form Solution to Photorealistic Image Stylization
- Shift-Net: Image Inpainting via Deep Feature Rearrangement
- Semantic Hierarchy Emerges in Deep Generative Representations for Scene Synthesis
- Exemplar Guided Unsupervised Image-to-Image Translation with Semantic Consistency
- InversionNet3D: Efficient and Scalable Learning for 3D Full Waveform Inversion
- Structural inpainting
- Visual Attribute Transfer through Deep Image Analogy
- Towards Metamerism via Foveated Style Transfer
- Bottleneck Supervised U-Net for Pixel-wise Liver and Tumor Segmentation
- Example-Guided Style Consistent Image Synthesis from Semantic Labeling
- Disentangling Multiple Conditional Inputs in GANs
- A Style-Aware Content Loss for Real-time HD Style Transfer
- ReCoNet: Real-time Coherent Video Style Transfer Network
- Quantitative Evaluation of Style Transfer
- Language Style Transfer from Sentences with Arbitrary Unknown Styles
- Towards Identifying and Managing Sources of Uncertainty in AI and Machine Learning Models - An Overview
- Judge the Judges: A Large-Scale Evaluation Study of Neural Language Models for Online Review Generation
- On the Diversity of Realistic Image Synthesis
- Neural Abstract Style Transfer for Chinese Traditional Painting
- GLStyleNet: Higher Quality Style Transfer Combining Global and Local Pyramid Features
- Im2Pencil: Controllable Pencil Illustration from Photographs
- Domain-Specific Mappings for Generative Adversarial Style Transfer
- A Flexible Convolutional Solver with Application to Photorealistic Style Transfer
- Data Analytics Service Composition and Deployment on Edge Devices
- Real-time Localized Photorealistic Video Style Transfer
- Uncorrelated Feature Encoding for Faster Image Style Transfer
- Learning Selfie-Friendly Abstraction from Artistic Style Images
- Paired 3D Model Generation with Conditional Generative Adversarial Networks
- Semantic Attribute Matching Networks
- Kunster -- AR Art Video Maker -- Real time video neural style transfer on mobile devices
- PortraitGAN for Flexible Portrait Manipulation
- CAMS: Color-Aware Multi-Style Transfer
- PSGAN++: Robust Detail-Preserving Makeup Transfer and Removal
- Deep Factorised Inverse-Sketching
- Non-Parametric Neural Style Transfer
- Learning to Transfer Visual Effects from Videos to Images
- Neural Style Difference Transfer and Its Application to Font Generation
- Synthesizing New Retinal Symptom Images by Multiple Generative Models