Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2603.11680 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908940533497856 |
|---|---|
| author | Tan, Cao Thien Trang, Phan Thi Thu Duc, Do Nghiem Anh, Ho Ngoc Zhuang, Hanyang Dung, Nguyen Duc |
| author_facet | Tan, Cao Thien Trang, Phan Thi Thu Duc, Do Nghiem Anh, Ho Ngoc Zhuang, Hanyang Dung, Nguyen Duc |
| contents | Hybrid CNN-Transformer architectures achieve strong results in image super-resolution, but scaling attention windows or convolution kernels significantly increases computational cost, limiting deployment on resource-constrained devices. We present UCAN, a lightweight network that unifies convolution and attention to expand the effective receptive field efficiently. UCAN combines window-based spatial attention with a Hedgehog Attention mechanism to model both local texture and long-range dependencies, and introduces a distillation-based large-kernel module to preserve high-frequency structure without heavy computation. In addition, we employ cross-layer parameter sharing to further reduce complexity. On Manga109 ($4\times$), UCAN-L achieves 31.63 dB PSNR with only 48.4G MACs, surpassing recent lightweight models. On BSDS100, UCAN attains 27.79 dB, outperforming methods with significantly larger models. Extensive experiments show that UCAN achieves a superior trade-off between accuracy, efficiency, and scalability, making it well-suited for practical high-resolution image restoration. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_11680 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | UCAN: Unified Convolutional Attention Network for Expansive Receptive Fields in Lightweight Super-Resolution Tan, Cao Thien Trang, Phan Thi Thu Duc, Do Nghiem Anh, Ho Ngoc Zhuang, Hanyang Dung, Nguyen Duc Computer Vision and Pattern Recognition Hybrid CNN-Transformer architectures achieve strong results in image super-resolution, but scaling attention windows or convolution kernels significantly increases computational cost, limiting deployment on resource-constrained devices. We present UCAN, a lightweight network that unifies convolution and attention to expand the effective receptive field efficiently. UCAN combines window-based spatial attention with a Hedgehog Attention mechanism to model both local texture and long-range dependencies, and introduces a distillation-based large-kernel module to preserve high-frequency structure without heavy computation. In addition, we employ cross-layer parameter sharing to further reduce complexity. On Manga109 ($4\times$), UCAN-L achieves 31.63 dB PSNR with only 48.4G MACs, surpassing recent lightweight models. On BSDS100, UCAN attains 27.79 dB, outperforming methods with significantly larger models. Extensive experiments show that UCAN achieves a superior trade-off between accuracy, efficiency, and scalability, making it well-suited for practical high-resolution image restoration. |
| title | UCAN: Unified Convolutional Attention Network for Expansive Receptive Fields in Lightweight Super-Resolution |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2603.11680 |