A diffusion probabilistic model for traditional Chinese landscape painting super-resolution
Abstract Traditional Chinese landscape painting is prone to low-resolution image issues during the digital protection process. To reconstruct high-quality images from low-resolution landscape paintings, we propose a novel Chinese landscape painting generation diffusion probabilistic model (CLDiff),...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SpringerOpen
2024-01-01
|
| Series: | Heritage Science |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s40494-023-01123-y |
| _version_ | 1827388283061010432 |
|---|---|
| author | Qiongshuai Lyu Na Zhao Yu Yang Yuehong Gong Jingli Gao |
| author_facet | Qiongshuai Lyu Na Zhao Yu Yang Yuehong Gong Jingli Gao |
| author_sort | Qiongshuai Lyu |
| collection | DOAJ |
| description | Abstract Traditional Chinese landscape painting is prone to low-resolution image issues during the digital protection process. To reconstruct high-quality images from low-resolution landscape paintings, we propose a novel Chinese landscape painting generation diffusion probabilistic model (CLDiff), which is similar to the Langevin dynamic process, and realizes the transformation of the Gaussian distribution into the empirical data distribution through multiple iterative refinement steps. The proposed CLDiff can provide ink texture clear super-resolution predictions by gradually transforming the pure Gaussian noise into a super-resolution landscape painting condition on a low-resolution input through a parameterized Markov Chain. Moreover, by introducing an attention module with an energy function into the U-Net architecture, we turn the denoising diffusion probabilistic model into a powerful generator. Experimental results show that CLDiff achieves better visual results and highly competitive performance in traditional Chinese Landscape painting super-resolution tasks. |
| first_indexed | 2024-03-08T16:15:28Z |
| format | Article |
| id | doaj.art-9ecdb88374384e6eaa6a28122f910f3a |
| institution | Directory Open Access Journal |
| issn | 2050-7445 |
| language | English |
| last_indexed | 2024-03-08T16:15:28Z |
| publishDate | 2024-01-01 |
| publisher | SpringerOpen |
| record_format | Article |
| series | Heritage Science |
| spelling | doaj.art-9ecdb88374384e6eaa6a28122f910f3a2024-01-07T12:38:53ZengSpringerOpenHeritage Science2050-74452024-01-0112111210.1186/s40494-023-01123-yA diffusion probabilistic model for traditional Chinese landscape painting super-resolutionQiongshuai Lyu0Na Zhao1Yu Yang2Yuehong Gong3Jingli Gao4School of Software, Pingdingshan UniversitySchool of Journalism and Communication, Pingdingshan UniversitySchool of Software, Pingdingshan UniversitySchool of Software, Pingdingshan UniversitySchool of Software, Pingdingshan UniversityAbstract Traditional Chinese landscape painting is prone to low-resolution image issues during the digital protection process. To reconstruct high-quality images from low-resolution landscape paintings, we propose a novel Chinese landscape painting generation diffusion probabilistic model (CLDiff), which is similar to the Langevin dynamic process, and realizes the transformation of the Gaussian distribution into the empirical data distribution through multiple iterative refinement steps. The proposed CLDiff can provide ink texture clear super-resolution predictions by gradually transforming the pure Gaussian noise into a super-resolution landscape painting condition on a low-resolution input through a parameterized Markov Chain. Moreover, by introducing an attention module with an energy function into the U-Net architecture, we turn the denoising diffusion probabilistic model into a powerful generator. Experimental results show that CLDiff achieves better visual results and highly competitive performance in traditional Chinese Landscape painting super-resolution tasks.https://doi.org/10.1186/s40494-023-01123-yChinese landscape paintingDenoising diffusion probabilistic modelAttention mechanismU-NetSuper-resolution |
| spellingShingle | Qiongshuai Lyu Na Zhao Yu Yang Yuehong Gong Jingli Gao A diffusion probabilistic model for traditional Chinese landscape painting super-resolution Heritage Science Chinese landscape painting Denoising diffusion probabilistic model Attention mechanism U-Net Super-resolution |
| title | A diffusion probabilistic model for traditional Chinese landscape painting super-resolution |
| title_full | A diffusion probabilistic model for traditional Chinese landscape painting super-resolution |
| title_fullStr | A diffusion probabilistic model for traditional Chinese landscape painting super-resolution |
| title_full_unstemmed | A diffusion probabilistic model for traditional Chinese landscape painting super-resolution |
| title_short | A diffusion probabilistic model for traditional Chinese landscape painting super-resolution |
| title_sort | diffusion probabilistic model for traditional chinese landscape painting super resolution |
| topic | Chinese landscape painting Denoising diffusion probabilistic model Attention mechanism U-Net Super-resolution |
| url | https://doi.org/10.1186/s40494-023-01123-y |
| work_keys_str_mv | AT qiongshuailyu adiffusionprobabilisticmodelfortraditionalchineselandscapepaintingsuperresolution AT nazhao adiffusionprobabilisticmodelfortraditionalchineselandscapepaintingsuperresolution AT yuyang adiffusionprobabilisticmodelfortraditionalchineselandscapepaintingsuperresolution AT yuehonggong adiffusionprobabilisticmodelfortraditionalchineselandscapepaintingsuperresolution AT jingligao adiffusionprobabilisticmodelfortraditionalchineselandscapepaintingsuperresolution AT qiongshuailyu diffusionprobabilisticmodelfortraditionalchineselandscapepaintingsuperresolution AT nazhao diffusionprobabilisticmodelfortraditionalchineselandscapepaintingsuperresolution AT yuyang diffusionprobabilisticmodelfortraditionalchineselandscapepaintingsuperresolution AT yuehonggong diffusionprobabilisticmodelfortraditionalchineselandscapepaintingsuperresolution AT jingligao diffusionprobabilisticmodelfortraditionalchineselandscapepaintingsuperresolution |