In the light of feature distributions: moment matching for Neural Style Transfer

Kalischek, Nikolai; Wegner, Jan Dirk; Schindler, Konrad

doi:10.1109/cvpr46437.2021.00926

Cited by 46 publications

(17 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Gatys also investigated what happens to generated images as a result of perceptual factors during style transfer as well as what happens to generated images afterward. e application of stylized images in real-world scenarios is severely constrained by the sluggishness of iterative optimization processes [10][11][12][13] due to the excessively slow rate at which stylized images are generated.…”

Section: Image Iteration and Generation Methodsmentioning

confidence: 99%

“…e loss function for this method, which trains an image modification model on a huge dataset and creates stylized images in the forward pass, can be defined using perceptual loss or Markov random field loss, depending on the circumstances. It is possible to transfer the style of images generated by generative adversarial networks (GANs) such as CycleGAN and DualGAN with great success [9][10][11][12][13][14]. But the dataset requirements and discriminative algorithms for GAN-based approaches are extremely rigorous.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Art Style Transfer of Oil Painting Based on Parallel Convolutional Neural Network

Jin

2022

Security and Communication Networks

View full text Add to dashboard Cite

To generate a new ornamental image, add an image’s oil painting style information to any image while preserving the image’s semantic content. With the rapid advancement of deep learning (DL), image style transfer has become one of the most active areas of computer vision research (CV). This paper proposes an oil painting style transfer technique based on parallel convolutional neural networks to address the ineffective style transfer of locally similar regions in content images and the slow processing speed of existing methods. By incorporating Gaussian sampling and a parallelization algorithm, this method effectively transfers the style of an oil painting. The algorithm can combine the content of any image with a variety of well-known oil painting styles to create high-quality works of art. The experimental results indicate that, compared to existing methods, the proposed method can effectively reduce the style loss of the generated image, make the generated image’s overall style more uniform, and produce a more pleasing visual effect.

show abstract

Section: Image Iteration and Generation Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Art Style Transfer of Oil Painting Based on Parallel Convolutional Neural Network

Jin

2022

Security and Communication Networks

View full text Add to dashboard Cite

show abstract

“…NST, first introduced in Gatys et al [21], can be optimization-based [9,21,22] or modelbased [16,28,30,51]. It is inherently defined in the image domain by matching CNN features either globally or in a local, patch-based manner [21,31,34,36,41]. Thus, it cannot directly utilize 3D data like depth or surface normals of a mesh.…”

Section: Related Workmentioning

confidence: 99%

StyleMesh: Style Transfer for Indoor 3D Scene Reconstructions

Höllein¹,

Johnson²,

Nießner³

2021

Preprint

View full text Add to dashboard Cite

Figure 1. We perform style transfer on reconstructed 3D meshes by synthesizing stylized textures. We compute style transfer losses on views of the scene and backpropagate gradients to the texture. Depth and surface normal data from the mesh enable 3D-aware stylization, preventing artifacts that arise from standard 2D losses. Our stylized meshes can be rendered using the traditional graphics pipeline.

show abstract

“…Inspired by previous works, several studies have synthesized realistic images using the properties of feature maps (Kalischek, Wegner, and Schindler 2021; Heitz et al 2021;Xu et al 2021). Many style transfer algorithms (Gatys, Ecker, and Bethge 2016;Lin et al 2021;Kalischek, Wegner, and Schindler 2021) extract the multi-scale feature maps on multiple layers by forwarding the style image to the network. Then they render the style from an image to a content image while preserving the semantic information of the content image.…”

Section: Introductionmentioning

confidence: 99%

NaturalInversion: Data-Free Image Synthesis Improving Real-World Consistency

Kim

Park

Kim

et al. 2022

AAAI

View full text Add to dashboard Cite

We introduce NaturalInversion, a novel model inversion-based method to synthesize images that agrees well with the original data distribution without using real data. In NaturalInversion, we propose: (1) a Feature Transfer Pyramid which uses enhanced image prior of the original data by combining the multi-scale feature maps extracted from the pre-trained classifier, (2) a one-to-one approach generative model where only one batch of images are synthesized by one generator to bring the non-linearity to optimization and to ease the overall optimizing process, (3) learnable Adaptive Channel Scaling parameters which are end-to-end trained to scale the output image channel to utilize the original image prior further. With our NaturalInversion, we synthesize images from classifiers trained on CIFAR-10/100 and show that our images are more consistent with original data distribution than prior works by visualization and additional analysis. Furthermore, our synthesized images outperform prior works on various applications such as knowledge distillation and pruning, demonstrating the effectiveness of our proposed method.

show abstract

In the light of feature distributions: moment matching for Neural Style Transfer

Cited by 46 publications

References 19 publications

Art Style Transfer of Oil Painting Based on Parallel Convolutional Neural Network

Art Style Transfer of Oil Painting Based on Parallel Convolutional Neural Network

StyleMesh: Style Transfer for Indoor 3D Scene Reconstructions

NaturalInversion: Data-Free Image Synthesis Improving Real-World Consistency

Contact Info

Product

Resources

About