Research map / Image generation / Super-resolution
PlugNet: Degradation Aware Scene Text Recognition Supervised by a Pluggable Super-Resolution Unit
Yongqiang Mou, Lei Tan, Hui Yang, Jingying Chen, Leyuan Liu, Rui Yan, Yaohong Huang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add Feature Enhance Module (FEM) (ablation) | PlugNet with FSM · SVT | accuracy: 90.0 -> 90.6 | Improved |
| add Feature Enhance Module (FEM) (ablation) | PlugNet with FSM · SVTP | accuracy: 80.8 -> 81.6 | Improved |
| add Feature Enhance Module (FEM) (ablation) | PlugNet with FSM · IC15 | accuracy: 78.2 -> 80.2 | Improved |
| add Feature Enhance Module (FEM) (ablation) | PlugNet with FSM · CUTE80 | accuracy: 82.6 -> 83.7 | Improved |
| add Pluggable Super-Resolution Unit (PSU) (ablation) | PlugNet · SVT | accuracy: 89.8 -> 92.3 | Improved |
| add Pluggable Super-Resolution Unit (PSU) (ablation) | PlugNet · SVTP | accuracy: 82.2 -> 84.3 | Improved |
| add Pluggable Super-Resolution Unit (PSU) (ablation) | PlugNet · IC15 | accuracy: 79.8 -> 82.6 | Improved |
| add Pluggable Super-Resolution Unit (PSU) (ablation) | PlugNet · CUTE80 | accuracy: 81.6 -> 85.0 | Improved |
| widen CNN feature resolution from 1x25 to 8x25 (FSM) (ablation) | PlugNet · SVT | accuracy: 85.2 -> 89.0 | Improved |
| use serial ESRGAN super-resolution instead of PSU (ablation) | PlugNet · SVT | accuracy: 89.8 -> 90.6 | No clear effect |
| set PSU loss weight lambda to 0.01 rather than larger values (ablation) | PlugNet · SVT, SVTP, IC15, CUTE80 | accuracy: 0.01 | Mixed |
Closest papers by meaning
- Gradient-Based Graph Attention for Scene Text Image Super-resolution (AAAI 2023 · similarity 0.86)
- Scene Text Telescope: Text-Focused Scene Image Super-Resolution (CVPR 2021 · similarity 0.85)
- Improving Scene Text Image Super-resolution via Dual Prior Modulation Network (AAAI 2023 · similarity 0.85)
- OTE: Exploring Accurate Scene Text Recognition Using One Token (CVPR 2024 · similarity 0.84)
- Unsupervised Degradation Representation Aware Transform for Real-World Blind Image Super-Resolution (AAAI 2025 · similarity 0.83)
- Improving Image Restoration through Removing Degradations in Textual Representations (CVPR 2024 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
