Application of Artificial Intelligence Virtual Image Technology in Photography Art Creation Under Deep Learning
With the continuous advancement of artificial intelligence (AI) and deep learning technologies, virtual image generation exhibits significant potential for application in photographic art creation. The primary objective of this study is to investigate the use of AI virtual image technology in photog...
Saved in:
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10840187/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832586850613592064 |
|---|---|
| author | Qiong Yao |
| author_facet | Qiong Yao |
| author_sort | Qiong Yao |
| collection | DOAJ |
| description | With the continuous advancement of artificial intelligence (AI) and deep learning technologies, virtual image generation exhibits significant potential for application in photographic art creation. The primary objective of this study is to investigate the use of AI virtual image technology in photography, particularly focusing on achieving creative expression and artistic style transfer through deep learning models. Consequently, this study proposes a novel model that integrates conditional generative adversarial networks (cGANs) with variational autoencoders (VAEs). This model aims to effectively address the challenges associated with image generation and style conversion in photographic art by leveraging the realistic generation capabilities of cGANs alongside the diversity maintenance features of VAEs. In the experimental section, the proposed cGANs + VAEs model is systematically compared with traditional Deep Convolutional GANs (DCGAN) and Pix2Pix models through empirical analysis. The experimental results indicate that the cGANs + VAEs model significantly outperforms traditional models in terms of image quality, artistic expression, and user satisfaction. Expert reviews further confirm the model’s superiority in artistic style imitation and creative generation. Additionally, user surveys reveal that most participants are highly satisfied with the images generated by the model, particularly regarding artistic perception and visual effects. Moreover, the cGANs + VAEs model demonstrates strong performance in Frechet Inception Distance (FID) and Inception Score (IS) across multiple datasets, yielding FID values of 13.67, 9.45, and 11.90 on the COCO, CelebA, and WikiArt datasets, respectively. In summary, the proposed cGANs + VAEs model not only achieves remarkable advancements in the technical performance of image generation but also exhibits considerable potential for practical applications in photographic art creation. |
| format | Article |
| id | doaj-art-3e92e0e04d904ee1865cd2731c825ca9 |
| institution | Kabale University |
| issn | 2169-3536 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-3e92e0e04d904ee1865cd2731c825ca92025-01-25T00:01:04ZengIEEEIEEE Access2169-35362025-01-0113145421455610.1109/ACCESS.2025.352952110840187Application of Artificial Intelligence Virtual Image Technology in Photography Art Creation Under Deep LearningQiong Yao0https://orcid.org/0009-0004-9027-4300Faculty of Fine Applied Arts and Cultural Sciences, Mahasarakham University, Maha Sarakham, ThailandWith the continuous advancement of artificial intelligence (AI) and deep learning technologies, virtual image generation exhibits significant potential for application in photographic art creation. The primary objective of this study is to investigate the use of AI virtual image technology in photography, particularly focusing on achieving creative expression and artistic style transfer through deep learning models. Consequently, this study proposes a novel model that integrates conditional generative adversarial networks (cGANs) with variational autoencoders (VAEs). This model aims to effectively address the challenges associated with image generation and style conversion in photographic art by leveraging the realistic generation capabilities of cGANs alongside the diversity maintenance features of VAEs. In the experimental section, the proposed cGANs + VAEs model is systematically compared with traditional Deep Convolutional GANs (DCGAN) and Pix2Pix models through empirical analysis. The experimental results indicate that the cGANs + VAEs model significantly outperforms traditional models in terms of image quality, artistic expression, and user satisfaction. Expert reviews further confirm the model’s superiority in artistic style imitation and creative generation. Additionally, user surveys reveal that most participants are highly satisfied with the images generated by the model, particularly regarding artistic perception and visual effects. Moreover, the cGANs + VAEs model demonstrates strong performance in Frechet Inception Distance (FID) and Inception Score (IS) across multiple datasets, yielding FID values of 13.67, 9.45, and 11.90 on the COCO, CelebA, and WikiArt datasets, respectively. In summary, the proposed cGANs + VAEs model not only achieves remarkable advancements in the technical performance of image generation but also exhibits considerable potential for practical applications in photographic art creation.https://ieeexplore.ieee.org/document/10840187/Deep learningconditional generative adversarial networksvariational autoencoderphotography artistic creationvirtual image technology |
| spellingShingle | Qiong Yao Application of Artificial Intelligence Virtual Image Technology in Photography Art Creation Under Deep Learning IEEE Access Deep learning conditional generative adversarial networks variational autoencoder photography artistic creation virtual image technology |
| title | Application of Artificial Intelligence Virtual Image Technology in Photography Art Creation Under Deep Learning |
| title_full | Application of Artificial Intelligence Virtual Image Technology in Photography Art Creation Under Deep Learning |
| title_fullStr | Application of Artificial Intelligence Virtual Image Technology in Photography Art Creation Under Deep Learning |
| title_full_unstemmed | Application of Artificial Intelligence Virtual Image Technology in Photography Art Creation Under Deep Learning |
| title_short | Application of Artificial Intelligence Virtual Image Technology in Photography Art Creation Under Deep Learning |
| title_sort | application of artificial intelligence virtual image technology in photography art creation under deep learning |
| topic | Deep learning conditional generative adversarial networks variational autoencoder photography artistic creation virtual image technology |
| url | https://ieeexplore.ieee.org/document/10840187/ |
| work_keys_str_mv | AT qiongyao applicationofartificialintelligencevirtualimagetechnologyinphotographyartcreationunderdeeplearning |