An improved defocusing adaptive style transfer method based on a stroke pyramid.
Image style transfer aims to assign a specified artist's style to a real image. However, most existing methods cannot generate textures of various thicknesses due to the rich semantic information of the input image. The image loses some semantic information through style transfer with a uniform...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Public Library of Science (PLoS)
2023-01-01
|
| Series: | PLoS ONE |
| Online Access: | https://doi.org/10.1371/journal.pone.0284742 |
| _version_ | 1827950536719073280 |
|---|---|
| author | Jianfang Cao Zeyu Chen Mengyan Jin Yun Tian |
| author_facet | Jianfang Cao Zeyu Chen Mengyan Jin Yun Tian |
| author_sort | Jianfang Cao |
| collection | DOAJ |
| description | Image style transfer aims to assign a specified artist's style to a real image. However, most existing methods cannot generate textures of various thicknesses due to the rich semantic information of the input image. The image loses some semantic information through style transfer with a uniform stroke size. To address the above problems, we propose an improved multi-stroke defocus adaptive style transfer framework based on a stroke pyramid, which mainly fuses various stroke sizes in the image spatial dimension to enhance the image content interpretability. We expand the receptive field of each branch and then fuse the features generated by the multiple branches based on defocus degree. Finally, we add an additional loss term to enhance the structural features of the generated image. The proposed model is trained using the Common Objects in Context (COCO) and Synthetic Depth of Field (SYNDOF) datasets, and the peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM) are used to evaluate the overall quality of the output image and its structural similarity with the content image, respectively. To validate the feasibility of the proposed algorithm, we compare the average PSNR and SSIM values of the output of the modified model and those of the original model. The experimental results show that the modified model improves the PSNR and SSIM values of the outputs by 1.43 and 0.12 on average, respectively. Compared with the single-stroke style transfer method, the framework proposed in this study improves the readability of the output images with more abundant visual expression. |
| first_indexed | 2024-04-09T13:28:15Z |
| format | Article |
| id | doaj.art-27d7a7d63b354c449aa489d9af6293e7 |
| institution | Directory Open Access Journal |
| issn | 1932-6203 |
| language | English |
| last_indexed | 2024-04-09T13:28:15Z |
| publishDate | 2023-01-01 |
| publisher | Public Library of Science (PLoS) |
| record_format | Article |
| series | PLoS ONE |
| spelling | doaj.art-27d7a7d63b354c449aa489d9af6293e72023-05-10T05:31:34ZengPublic Library of Science (PLoS)PLoS ONE1932-62032023-01-01184e028474210.1371/journal.pone.0284742An improved defocusing adaptive style transfer method based on a stroke pyramid.Jianfang CaoZeyu ChenMengyan JinYun TianImage style transfer aims to assign a specified artist's style to a real image. However, most existing methods cannot generate textures of various thicknesses due to the rich semantic information of the input image. The image loses some semantic information through style transfer with a uniform stroke size. To address the above problems, we propose an improved multi-stroke defocus adaptive style transfer framework based on a stroke pyramid, which mainly fuses various stroke sizes in the image spatial dimension to enhance the image content interpretability. We expand the receptive field of each branch and then fuse the features generated by the multiple branches based on defocus degree. Finally, we add an additional loss term to enhance the structural features of the generated image. The proposed model is trained using the Common Objects in Context (COCO) and Synthetic Depth of Field (SYNDOF) datasets, and the peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM) are used to evaluate the overall quality of the output image and its structural similarity with the content image, respectively. To validate the feasibility of the proposed algorithm, we compare the average PSNR and SSIM values of the output of the modified model and those of the original model. The experimental results show that the modified model improves the PSNR and SSIM values of the outputs by 1.43 and 0.12 on average, respectively. Compared with the single-stroke style transfer method, the framework proposed in this study improves the readability of the output images with more abundant visual expression.https://doi.org/10.1371/journal.pone.0284742 |
| spellingShingle | Jianfang Cao Zeyu Chen Mengyan Jin Yun Tian An improved defocusing adaptive style transfer method based on a stroke pyramid. PLoS ONE |
| title | An improved defocusing adaptive style transfer method based on a stroke pyramid. |
| title_full | An improved defocusing adaptive style transfer method based on a stroke pyramid. |
| title_fullStr | An improved defocusing adaptive style transfer method based on a stroke pyramid. |
| title_full_unstemmed | An improved defocusing adaptive style transfer method based on a stroke pyramid. |
| title_short | An improved defocusing adaptive style transfer method based on a stroke pyramid. |
| title_sort | improved defocusing adaptive style transfer method based on a stroke pyramid |
| url | https://doi.org/10.1371/journal.pone.0284742 |
| work_keys_str_mv | AT jianfangcao animproveddefocusingadaptivestyletransfermethodbasedonastrokepyramid AT zeyuchen animproveddefocusingadaptivestyletransfermethodbasedonastrokepyramid AT mengyanjin animproveddefocusingadaptivestyletransfermethodbasedonastrokepyramid AT yuntian animproveddefocusingadaptivestyletransfermethodbasedonastrokepyramid AT jianfangcao improveddefocusingadaptivestyletransfermethodbasedonastrokepyramid AT zeyuchen improveddefocusingadaptivestyletransfermethodbasedonastrokepyramid AT mengyanjin improveddefocusingadaptivestyletransfermethodbasedonastrokepyramid AT yuntian improveddefocusingadaptivestyletransfermethodbasedonastrokepyramid |