Detail-Preserving CycleGAN-AdaIN Framework for Image-to-Ink Painting Translation

Zhang, Fengquan; Gao, Huaming; Lai, Yuping

doi:10.1109/access.2020.3009470

Cited by 23 publications

(10 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…e texture tracking and matching method is used for the information fusion of works of art. Firstly, the feature analysis model (x, y, z) of works of art with color features is constructed [14].…”

Section: Color Feature Region Segmentation Of Work Of Artmentioning

confidence: 99%

Design of Artistic Creation Style Extraction Model Based on Color Feature Data

Yao

Sohail

2022

Mathematical Problems in Engineering

View full text Add to dashboard Cite

In order to improve the style extraction and extraction ability of works of art, a color extraction method based on color features is proposed. The color feature extraction method is used to extract the style of visual works of art. The color feature region of works of art is segmented combined with a sparse scattered point reorganization method. Texture tracking and matching method is used for information fusion of works of art, combined with corner detection, three-dimensional edge contour feature detection method to realize texture filling and automatic rendering of art color extraction to improve art graphics’ color visual feature expression ability. The color feature data method is used for visual feature sampling and equalizing works of art. According to the equilibrium configuration results, the fuzzy clustering method is used to extract the color style of works of art to improve the style extraction and extraction identification ability of works of art. The simulation results show that this method has high accuracy in color extraction of works of art. It has a good effect on the extraction of art creation style and improves the ability of three-dimensional extraction and automatic extraction of art.

show abstract

Section: Color Feature Region Segmentation Of Work Of Artmentioning

confidence: 99%

Design of Artistic Creation Style Extraction Model Based on Color Feature Data

Yao

Sohail

2022

Mathematical Problems in Engineering

View full text Add to dashboard Cite

show abstract

“…And dynamic layer normalization (DLN) [25] relies on a neural network to generate the β and γ . And similar to the method adopted in the works [26], [27], [28], [29], VOLUME 10, 2022 the generate parameters by using a dedicated neural network dynamically in this paper.…”

Section: B Cnn Basic Approachmentioning

confidence: 98%

A Generative Adversarial Network for Multistyle Unsupervised Image Transfer Based on Dynamic Position Normalization

Rang

Zheng

et al. 2022

IEEE Access

View full text Add to dashboard Cite

Image transfer is a technology that changes the image effect by processing the image's color, contour, line, and other information. To stylize the visual appearance of the output will be adapted to the subject of the original image, this paper proposes an image feature transfer method called Cycle-DPN-GAN, based on the Cycle-Consistent Adversarial Networks. Firstly, Positional Normalization-Dynamic Moment Shortcut (PONO-DMS) module is introduced to learn more structural information from the input image, and the edge blurring and object losing are efficiently alleviated. In addition, the Multi-Scale-Structural Similarity Index (MS-SSIM) loss is added to the reconstruction loss, which improves visual perceptions and enhances the constraints on the reconstructed image in terms of image brightness, color contrast and structure. In this model, to verify the feasibility and superiority of the proposed method, the data sets of monet2photo, vangogh2photo, ukiyoe2photo and cezanne2photo are performed in the experiments, and the Inception Score and Fréchet Inception Distance evaluation index are improved. In addition, ablation studies are performed to demonstrate the validity of each proposed component. In this paper, the results of the quantitative evaluation are consistent with the qualitative evaluation. It can be demonstrated that the images generated by Cycle-DPN-GAN have higher visual quality. achieves better performance in both specific quantitative and 49 qualitative evaluation indexes. In the comparation with other 50 networks (Fig. 1), our proposed network can preserve more 51 details of the original image when obtaining a much clearer 52 stylized image. The contributions of this paper are as follows: 53 1) The structural information of the input image is 54 extracted by PONO [6] before features transforma-55 tion and then transmitted into the DMS layer, which 56 are used to recombine the stylized image. By this 57 way, the edge blurring and object losing are efficiently 58 alleviated.59 2) In order to improve the brightness, color contrast, 60 and composition, cycle consistency loss and MS-SSIM 61 loss [7] are used jointly in the generative network to 62 strengthen the constraints for realistic and natural artist 63 painting. 64 The following of this paper is organized as: Section II 65 introduces the researches of image-to-image style trans-66 fer; Section III expounds the proposed Cycle-DPN-GAN; 67 Section IV states the details of implementation, shows and 68 analyzes the experimental results; Section V gives a brief 69 conclusion of this work. 70 II. RELATED WORK 71 In this section, the recent researches of image style transfer 72 are reviewed briefly, and the related style transformation 73 methods are divided into three categories: Texture synthesis, 74 CNN based approaches and GAN based approaches.

show abstract

“…A CycleGAN-AdaIN framework is proposed to emphasize the details of generated Chinese paintings [15]. An AdaIN structure is added between the encoder and decoder of generators in the CycleGAN model to mimic the blur of real ink wash painting, and the cycle constraint between the generated painting and reconstructed photo is cancelled to encourage the model to be more daring in void-leaving and edge-emphasizing [3, 15~16].…”

Section: Models Specifically For Generating Chinese Paintingsmentioning

confidence: 99%

Improvements on style transfer from photographs to Chinese ink wash paintings based on generative adversarial networks

Zhang¹

2023

Third International Conference on Intelligent Computing and Human-Computer Interaction (ICHCI 2022)

View full text Add to dashboard Cite

Although voids play an important role in Chinese paintings, when generating Chinese ink wash paintings from real life photos of horses, background detection and void-leaving remain challenges in producing more credible paintings. To address the problem, a model with a two-stage framework is proposed in this paper. The framework divides the generation process into background lightening and style transformation. In the first stage, by training a pix2pix model with paired original photos and background-lightened photos, the pix2pix model is enabled to detect and lighten the background correctly in the style-transfer process, hinting where voids should be left. In the second stage, a Cycle-GAN model trained with unmatched photos and Chinese paintings achieves the style-transfer from pre-processed photos to Chinese ink wash paintings. In preparation for training the stage-Ⅰ model, the photos from a given dataset is processed in batches to get corresponding background-lightened photos. During the training of the stage-Ⅱ model, original photos instead of processed photos is used as comparatively data augmentation, making the model more robust and independent. Compared to the baseline model, the proposed model reaches a higher accuracy in both void-leaving and detailenhancing in the experiments, resulting in more credible and delicate generated Chinese paintings.

show abstract

Detail-Preserving CycleGAN-AdaIN Framework for Image-to-Ink Painting Translation

Cited by 23 publications

References 19 publications

Design of Artistic Creation Style Extraction Model Based on Color Feature Data

Design of Artistic Creation Style Extraction Model Based on Color Feature Data

A Generative Adversarial Network for Multistyle Unsupervised Image Transfer Based on Dynamic Position Normalization

Improvements on style transfer from photographs to Chinese ink wash paintings based on generative adversarial networks

Contact Info

Product

Resources

About