3DAttGAN: A 3D Attention-based Generative Adversarial Network for Joint Space-Time Video Super-Resolution
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866913443716530176 |
|---|---|
| author | Fu, Congrui Yuan, Hui Shen, Liquan Hamzaoui, Raouf Zhang, Hao |
| author_facet | Fu, Congrui Yuan, Hui Shen, Liquan Hamzaoui, Raouf Zhang, Hao |
| contents | In many applications, including surveillance, entertainment, and restoration, there is a need to increase both the spatial resolution and the frame rate of a video sequence. The aim is to improve visual quality, refine details, and create a more realistic viewing experience. Existing space-time video super-resolution methods do not effectively use spatio-temporal information. To address this limitation, we propose a generative adversarial network for joint space-time video super-resolution. The generative network consists of three operations: shallow feature extraction, deep feature extraction, and reconstruction. It uses three-dimensional (3D) convolutions to process temporal and spatial information simultaneously and includes a novel 3D attention mechanism to extract the most important channel and spatial information. The discriminative network uses a two-branch structure to handle details and motion information, making the generated results more accurate. Experimental results on the Vid4, Vimeo-90K, and REDS datasets demonstrate the effectiveness of the proposed method. The source code is publicly available at https://github.com/FCongRui/3DAttGan.git. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_16965 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | 3DAttGAN: A 3D Attention-based Generative Adversarial Network for Joint Space-Time Video Super-Resolution Fu, Congrui Yuan, Hui Shen, Liquan Hamzaoui, Raouf Zhang, Hao Image and Video Processing In many applications, including surveillance, entertainment, and restoration, there is a need to increase both the spatial resolution and the frame rate of a video sequence. The aim is to improve visual quality, refine details, and create a more realistic viewing experience. Existing space-time video super-resolution methods do not effectively use spatio-temporal information. To address this limitation, we propose a generative adversarial network for joint space-time video super-resolution. The generative network consists of three operations: shallow feature extraction, deep feature extraction, and reconstruction. It uses three-dimensional (3D) convolutions to process temporal and spatial information simultaneously and includes a novel 3D attention mechanism to extract the most important channel and spatial information. The discriminative network uses a two-branch structure to handle details and motion information, making the generated results more accurate. Experimental results on the Vid4, Vimeo-90K, and REDS datasets demonstrate the effectiveness of the proposed method. The source code is publicly available at https://github.com/FCongRui/3DAttGan.git. |
| title | 3DAttGAN: A 3D Attention-based Generative Adversarial Network for Joint Space-Time Video Super-Resolution |
| topic | Image and Video Processing |
| url | https://arxiv.org/abs/2407.16965 |