Research map / Detection and segmentation / CNN features
MOST: A Multi-Oriented Scene Text Detector With Localization Refinement
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 TFAM to the EAST baseline (ablation) | EAST text detector · MLT17 val IoU@0.7 | F: 61.4 -> 68.8 | Helped |
| Add TFAM to the EAST baseline (ablation) | EAST text detector · MTWI test | F: 69.1 -> 73.3 | Helped |
| Add instance-wise IoU loss to the EAST baseline (ablation) | EAST text detector · MLT17 val IoU@0.5 | F: 75.0 -> 76.0 | Helped |
| Add PA-NMS to the EAST baseline (ablation) | EAST text detector · MTWI test | F: 69.1 -> 71.8 | Helped |
| Add instance-wise IoU loss to TFAM (ablation) | EAST with TFAM · MLT17 val IoU@0.5 | F: 77.4 -> 78.5 | Helped |
| Add PA-NMS to TFAM and instance-wise IoU loss (ablation) | EAST with TFAM and instance-wise IoU loss · MTWI test | F: 73.8 -> 74.2 | Helped |
| Use feature-based sampling in TFAM instead of no TFAM (ablation) | EAST text detector · MLT17 val IoU@0.5 | F: 75.0 -> 75.8 | Helped |
| Use localization-based sampling in TFAM instead of no TFAM (ablation) | EAST text detector · MLT17 val IoU@0.5 | F: 75.0 -> 77.3 | Helped |
| Combine feature-based and localization-based sampling in TFAM (ablation) | EAST text detector · MLT17 val IoU@0.5 | F: 75.0 -> 77.4 | Helped |
| Combine feature-based and localization-based sampling instead of localization-based sampling alone (ablation) | TFAM · MLT17 val IoU@0.7 | F: 68.0 -> 68.8 | Helped |
| Add all three strategies: TFAM, instance-wise IoU loss, and PA-NMS (ablation) | EAST text detector · MLT17 val IoU@0.5 | F: 75.0 -> 79.0 | Helped |
| Add all three strategies: TFAM, instance-wise IoU loss, and PA-NMS (ablation) | EAST text detector · MLT17 val IoU@0.7 | F: 61.4 -> 70.9 | Helped |
Closest papers by meaning
- Multi-Oriented Scene Text Detection via Corner Localization and Region Segmentation (CVPR 2018 · similarity 0.90)
- Canny Text Detector: Fast and Robust Scene Text Localization Algorithm (CVPR 2016 · similarity 0.89)
- Text Grouping Adapter: Adapting Pre-trained Text Detector for Layout Analysis (CVPR 2024 · similarity 0.86)
- Arbitrary Reading Order Scene Text Spotter with Local Semantics Guidance (AAAI 2025 · similarity 0.86)
- Kernel Adaptive Convolution for Scene Text Detection via Distance Map Prediction (CVPR 2024 · similarity 0.86)
- LayoutFormer: Hierarchical Text Detection Towards Scene Text Understanding (CVPR 2024 · similarity 0.85)
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.