Controlling Neural Style Transfer with Deep Reinforcement Learning
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2023
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866917747164708864 |
|---|---|
| author | Feng, Chengming Hu, Jing Wang, Xin Hu, Shu Zhu, Bin Wu, Xi Zhu, Hongtu Lyu, Siwei |
| author_facet | Feng, Chengming Hu, Jing Wang, Xin Hu, Shu Zhu, Bin Wu, Xi Zhu, Hongtu Lyu, Siwei |
| contents | Controlling the degree of stylization in the Neural Style Transfer (NST) is a little tricky since it usually needs hand-engineering on hyper-parameters. In this paper, we propose the first deep Reinforcement Learning (RL) based architecture that splits one-step style transfer into a step-wise process for the NST task. Our RL-based method tends to preserve more details and structures of the content image in early steps, and synthesize more style patterns in later steps. It is a user-easily-controlled style-transfer method. Additionally, as our RL-based model performs the stylization progressively, it is lightweight and has lower computational complexity than existing one-step Deep Learning (DL) based models. Experimental results demonstrate the effectiveness and robustness of our method. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2310_00405 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Controlling Neural Style Transfer with Deep Reinforcement Learning Feng, Chengming Hu, Jing Wang, Xin Hu, Shu Zhu, Bin Wu, Xi Zhu, Hongtu Lyu, Siwei Computer Vision and Pattern Recognition Controlling the degree of stylization in the Neural Style Transfer (NST) is a little tricky since it usually needs hand-engineering on hyper-parameters. In this paper, we propose the first deep Reinforcement Learning (RL) based architecture that splits one-step style transfer into a step-wise process for the NST task. Our RL-based method tends to preserve more details and structures of the content image in early steps, and synthesize more style patterns in later steps. It is a user-easily-controlled style-transfer method. Additionally, as our RL-based model performs the stylization progressively, it is lightweight and has lower computational complexity than existing one-step Deep Learning (DL) based models. Experimental results demonstrate the effectiveness and robustness of our method. |
| title | Controlling Neural Style Transfer with Deep Reinforcement Learning |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2310.00405 |