CU-Mamba: Selective State Space Models with Channel Learning 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_ | 1866913319911161856 |
|---|---|
| author | Deng, Rui Gu, Tianpei |
| author_facet | Deng, Rui Gu, Tianpei |
| contents | Reconstructing degraded images is a critical task in image processing. Although CNN and Transformer-based models are prevalent in this field, they exhibit inherent limitations, such as inadequate long-range dependency modeling and high computational costs. To overcome these issues, we introduce the Channel-Aware U-Shaped Mamba (CU-Mamba) model, which incorporates a dual State Space Model (SSM) framework into the U-Net architecture. CU-Mamba employs a Spatial SSM module for global context encoding and a Channel SSM component to preserve channel correlation features, both in linear computational complexity relative to the feature map size. Extensive experimental results validate CU-Mamba's superiority over existing state-of-the-art methods, underscoring the importance of integrating both spatial and channel contexts in image restoration. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_11778 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | CU-Mamba: Selective State Space Models with Channel Learning for Image Restoration Deng, Rui Gu, Tianpei Computer Vision and Pattern Recognition Reconstructing degraded images is a critical task in image processing. Although CNN and Transformer-based models are prevalent in this field, they exhibit inherent limitations, such as inadequate long-range dependency modeling and high computational costs. To overcome these issues, we introduce the Channel-Aware U-Shaped Mamba (CU-Mamba) model, which incorporates a dual State Space Model (SSM) framework into the U-Net architecture. CU-Mamba employs a Spatial SSM module for global context encoding and a Channel SSM component to preserve channel correlation features, both in linear computational complexity relative to the feature map size. Extensive experimental results validate CU-Mamba's superiority over existing state-of-the-art methods, underscoring the importance of integrating both spatial and channel contexts in image restoration. |
| title | CU-Mamba: Selective State Space Models with Channel Learning for Image Restoration |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2404.11778 |