Attention-assisted dual-branch interactive face super-resolution network
We propose a deep learning-based Attention-Assisted Dual-Branch Interactive Network (ADBINet) to improve facial super-resolution by addressing key challenges like inadequate feature extraction and poor multi-scale information handling. ADBINet features a multi-scale encoder-decoder architecture that...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
KeAi Communications Co. Ltd.
2025-01-01
|
| Series: | Cognitive Robotics |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2667241325000023 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832582091020173312 |
|---|---|
| author | Xujie Wan Siyu Xu Guangwei Gao |
| author_facet | Xujie Wan Siyu Xu Guangwei Gao |
| author_sort | Xujie Wan |
| collection | DOAJ |
| description | We propose a deep learning-based Attention-Assisted Dual-Branch Interactive Network (ADBINet) to improve facial super-resolution by addressing key challenges like inadequate feature extraction and poor multi-scale information handling. ADBINet features a multi-scale encoder-decoder architecture that captures and integrates features across scales, enhancing detail and reconstruction quality. The key to our approach is the Transformer and CNN Interaction Module (TCIM), which includes a Dual Attention Collaboration Module (DACM) for improved local and spatial feature extraction. The Channel Attention Guidance Module (CAGM) refines CNN and Transformer fusion, ensuring precise facial detail restoration. Additionally, the Attention Feature Fusion Unit (AFFM) optimizes multi-scale feature integration. Experimental results demonstrate that ADBINet outperforms existing methods in both quantitative and qualitative facial super-resolution metrics. |
| format | Article |
| id | doaj-art-5558f4623530409fa18f44d67ce54a0d |
| institution | Kabale University |
| issn | 2667-2413 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | KeAi Communications Co. Ltd. |
| record_format | Article |
| series | Cognitive Robotics |
| spelling | doaj-art-5558f4623530409fa18f44d67ce54a0d2025-01-30T05:15:11ZengKeAi Communications Co. Ltd.Cognitive Robotics2667-24132025-01-0157785Attention-assisted dual-branch interactive face super-resolution networkXujie Wan0Siyu Xu1Guangwei Gao2Institute of Advanced Technology, Nanjing University of Posts and Telecommunications, Nanjing, ChinaInstitute of Advanced Technology, Nanjing University of Posts and Telecommunications, Nanjing, ChinaInstitute of Advanced Technology, Nanjing University of Posts and Telecommunications, Nanjing, China; Provincial Key Laboratory for Computer Information Processing Technology, Soochow University, Suzhou, China; Corresponding author.We propose a deep learning-based Attention-Assisted Dual-Branch Interactive Network (ADBINet) to improve facial super-resolution by addressing key challenges like inadequate feature extraction and poor multi-scale information handling. ADBINet features a multi-scale encoder-decoder architecture that captures and integrates features across scales, enhancing detail and reconstruction quality. The key to our approach is the Transformer and CNN Interaction Module (TCIM), which includes a Dual Attention Collaboration Module (DACM) for improved local and spatial feature extraction. The Channel Attention Guidance Module (CAGM) refines CNN and Transformer fusion, ensuring precise facial detail restoration. Additionally, the Attention Feature Fusion Unit (AFFM) optimizes multi-scale feature integration. Experimental results demonstrate that ADBINet outperforms existing methods in both quantitative and qualitative facial super-resolution metrics.http://www.sciencedirect.com/science/article/pii/S266724132500002300-0199-00 |
| spellingShingle | Xujie Wan Siyu Xu Guangwei Gao Attention-assisted dual-branch interactive face super-resolution network Cognitive Robotics 00-01 99-00 |
| title | Attention-assisted dual-branch interactive face super-resolution network |
| title_full | Attention-assisted dual-branch interactive face super-resolution network |
| title_fullStr | Attention-assisted dual-branch interactive face super-resolution network |
| title_full_unstemmed | Attention-assisted dual-branch interactive face super-resolution network |
| title_short | Attention-assisted dual-branch interactive face super-resolution network |
| title_sort | attention assisted dual branch interactive face super resolution network |
| topic | 00-01 99-00 |
| url | http://www.sciencedirect.com/science/article/pii/S2667241325000023 |
| work_keys_str_mv | AT xujiewan attentionassisteddualbranchinteractivefacesuperresolutionnetwork AT siyuxu attentionassisteddualbranchinteractivefacesuperresolutionnetwork AT guangweigao attentionassisteddualbranchinteractivefacesuperresolutionnetwork |