CascadedGaze: Efficiency in Global Context Extraction for Image Restoration
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866913343865880576 |
|---|---|
| author | Ghasemabadi, Amirhosein Janjua, Muhammad Kamran Salameh, Mohammad Zhou, Chunhua Sun, Fengyu Niu, Di |
| author_facet | Ghasemabadi, Amirhosein Janjua, Muhammad Kamran Salameh, Mohammad Zhou, Chunhua Sun, Fengyu Niu, Di |
| contents | Image restoration tasks traditionally rely on convolutional neural networks. However, given the local nature of the convolutional operator, they struggle to capture global information. The promise of attention mechanisms in Transformers is to circumvent this problem, but it comes at the cost of intensive computational overhead. Many recent studies in image restoration have focused on solving the challenge of balancing performance and computational cost via Transformer variants. In this paper, we present CascadedGaze Network (CGNet), an encoder-decoder architecture that employs Global Context Extractor (GCE), a novel and efficient way to capture global information for image restoration. The GCE module leverages small kernels across convolutional layers to learn global dependencies, without requiring self-attention. Extensive experimental results show that our computationally efficient approach performs competitively to a range of state-of-the-art methods on synthetic image denoising and single image deblurring tasks, and pushes the performance boundary further on the real image denoising task. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2401_15235 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | CascadedGaze: Efficiency in Global Context Extraction for Image Restoration Ghasemabadi, Amirhosein Janjua, Muhammad Kamran Salameh, Mohammad Zhou, Chunhua Sun, Fengyu Niu, Di Image and Video Processing Computer Vision and Pattern Recognition Machine Learning Image restoration tasks traditionally rely on convolutional neural networks. However, given the local nature of the convolutional operator, they struggle to capture global information. The promise of attention mechanisms in Transformers is to circumvent this problem, but it comes at the cost of intensive computational overhead. Many recent studies in image restoration have focused on solving the challenge of balancing performance and computational cost via Transformer variants. In this paper, we present CascadedGaze Network (CGNet), an encoder-decoder architecture that employs Global Context Extractor (GCE), a novel and efficient way to capture global information for image restoration. The GCE module leverages small kernels across convolutional layers to learn global dependencies, without requiring self-attention. Extensive experimental results show that our computationally efficient approach performs competitively to a range of state-of-the-art methods on synthetic image denoising and single image deblurring tasks, and pushes the performance boundary further on the real image denoising task. |
| title | CascadedGaze: Efficiency in Global Context Extraction for Image Restoration |
| topic | Image and Video Processing Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2401.15235 |