Artistic characterization of AI painting based on generative adversarial networks
Combined with the creation process of AI painting art, it analyzes the artistic design characteristics of AI paintings formed by generative adversarial networks. It utilizes a convolutional neural network to extract the artistic characteristics of AI paintings and combines the error of feature loss...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Sciendo
2024-01-01
|
| Series: | Applied Mathematics and Nonlinear Sciences |
| Subjects: | |
| Online Access: | https://doi.org/10.2478/amns-2024-0238 |
| _version_ | 1797303125431812096 |
|---|---|
| author | Lu Weiwei Qi Ruixing Li Yuhui |
| author_facet | Lu Weiwei Qi Ruixing Li Yuhui |
| author_sort | Lu Weiwei |
| collection | DOAJ |
| description | Combined with the creation process of AI painting art, it analyzes the artistic design characteristics of AI paintings formed by generative adversarial networks. It utilizes a convolutional neural network to extract the artistic characteristics of AI paintings and combines the error of feature loss to calculate the features, which ensures the stable operation of the generative adversarial network model. To achieve the style migration of AI painting artworks, the Cycle GAN model was designed on this basis. Comparing the features of both AI paintings of generative adversarial networks and paintings of human artists, the perceptual complexity is taken as the dependent variable, and a regression model is established to analyze and calculate the complexity features of AI paintings, as well as to analyze the color matching art of AI paintings by combining the beauty calculation method. According to the comparison results, the AI paintings have a score of 3.71 for inspirational, 3.69 for aesthetic value, 3.52 for compositional rationality, and 3.38 for breakthrough. The AI paintings have a high level of thought and inspirational value. |
| first_indexed | 2024-03-07T23:48:26Z |
| format | Article |
| id | doaj.art-b3ed789f23fb4a36be8c196bfccf2fa1 |
| institution | Directory Open Access Journal |
| issn | 2444-8656 |
| language | English |
| last_indexed | 2024-03-07T23:48:26Z |
| publishDate | 2024-01-01 |
| publisher | Sciendo |
| record_format | Article |
| series | Applied Mathematics and Nonlinear Sciences |
| spelling | doaj.art-b3ed789f23fb4a36be8c196bfccf2fa12024-02-19T09:03:36ZengSciendoApplied Mathematics and Nonlinear Sciences2444-86562024-01-019110.2478/amns-2024-0238Artistic characterization of AI painting based on generative adversarial networksLu Weiwei0Qi Ruixing1Li Yuhui21School of Design, NingboTech University, Ningbo, Zhejiang, 315100, China.2Basic Teaching Department, Hebei Academy of Fine Arts, Shijiazhuang, Hebei, 050700, China.1School of Design, NingboTech University, Ningbo, Zhejiang, 315100, China.Combined with the creation process of AI painting art, it analyzes the artistic design characteristics of AI paintings formed by generative adversarial networks. It utilizes a convolutional neural network to extract the artistic characteristics of AI paintings and combines the error of feature loss to calculate the features, which ensures the stable operation of the generative adversarial network model. To achieve the style migration of AI painting artworks, the Cycle GAN model was designed on this basis. Comparing the features of both AI paintings of generative adversarial networks and paintings of human artists, the perceptual complexity is taken as the dependent variable, and a regression model is established to analyze and calculate the complexity features of AI paintings, as well as to analyze the color matching art of AI paintings by combining the beauty calculation method. According to the comparison results, the AI paintings have a score of 3.71 for inspirational, 3.69 for aesthetic value, 3.52 for compositional rationality, and 3.38 for breakthrough. The AI paintings have a high level of thought and inspirational value.https://doi.org/10.2478/amns-2024-0238generative adversarial networkconvolutional neural networkregression modelai painting05c82 |
| spellingShingle | Lu Weiwei Qi Ruixing Li Yuhui Artistic characterization of AI painting based on generative adversarial networks Applied Mathematics and Nonlinear Sciences generative adversarial network convolutional neural network regression model ai painting 05c82 |
| title | Artistic characterization of AI painting based on generative adversarial networks |
| title_full | Artistic characterization of AI painting based on generative adversarial networks |
| title_fullStr | Artistic characterization of AI painting based on generative adversarial networks |
| title_full_unstemmed | Artistic characterization of AI painting based on generative adversarial networks |
| title_short | Artistic characterization of AI painting based on generative adversarial networks |
| title_sort | artistic characterization of ai painting based on generative adversarial networks |
| topic | generative adversarial network convolutional neural network regression model ai painting 05c82 |
| url | https://doi.org/10.2478/amns-2024-0238 |
| work_keys_str_mv | AT luweiwei artisticcharacterizationofaipaintingbasedongenerativeadversarialnetworks AT qiruixing artisticcharacterizationofaipaintingbasedongenerativeadversarialnetworks AT liyuhui artisticcharacterizationofaipaintingbasedongenerativeadversarialnetworks |