FADPNet: Frequency-Aware Dual-Path Network for Face Super-Resolution
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917412400529408 |
|---|---|
| author | Xu, Siyu Li, Wenjie Gao, Guangwei Yang, Jian Qi, Guo-Jun Lin, Chia-Wen |
| author_facet | Xu, Siyu Li, Wenjie Gao, Guangwei Yang, Jian Qi, Guo-Jun Lin, Chia-Wen |
| contents | Face super-resolution (FSR) under limited computational budgets remains challenging. Existing methods often treat all facial pixels equally, leading to suboptimal resource allocation and degraded performance. CNNs are sensitive to high-frequency facial features such as contours and outlines, while Mamba excels at capturing low-frequency attributes like facial color and texture with lower complexity than Transformers. Motivated by this, we propose FADPNet, a Frequency-Aware Dual-Path Network that decomposes facial features into low- and high-frequency components for dedicated processing. The low-frequency branch employs a Mamba-based Low-Frequency Enhancement Block (LFEB) that integrates state-space attention with squeeze-and-excitation to restore global interactions and emphasize informative channels. The high-frequency branch uses a CNN-based Deep Position-Aware Attention (DPA) module to refine structural details, followed by a lightweight High-Frequency Refinement (HFR) module for further frequency-specific refinement. These designs enable FADPNet to achieve a strong balance between FSR quality and efficiency, outperforming existing methods. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_14121 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | FADPNet: Frequency-Aware Dual-Path Network for Face Super-Resolution Xu, Siyu Li, Wenjie Gao, Guangwei Yang, Jian Qi, Guo-Jun Lin, Chia-Wen Computer Vision and Pattern Recognition Face super-resolution (FSR) under limited computational budgets remains challenging. Existing methods often treat all facial pixels equally, leading to suboptimal resource allocation and degraded performance. CNNs are sensitive to high-frequency facial features such as contours and outlines, while Mamba excels at capturing low-frequency attributes like facial color and texture with lower complexity than Transformers. Motivated by this, we propose FADPNet, a Frequency-Aware Dual-Path Network that decomposes facial features into low- and high-frequency components for dedicated processing. The low-frequency branch employs a Mamba-based Low-Frequency Enhancement Block (LFEB) that integrates state-space attention with squeeze-and-excitation to restore global interactions and emphasize informative channels. The high-frequency branch uses a CNN-based Deep Position-Aware Attention (DPA) module to refine structural details, followed by a lightweight High-Frequency Refinement (HFR) module for further frequency-specific refinement. These designs enable FADPNet to achieve a strong balance between FSR quality and efficiency, outperforming existing methods. |
| title | FADPNet: Frequency-Aware Dual-Path Network for Face Super-Resolution |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2506.14121 |