Research map / Detection and segmentation / CNN features
Towards Robust Curve Text Detection With Conditional Spatial Expansion
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 |
|---|---|---|---|
| Use conditional spatial expansion instead of Mask R-CNN segmentation | Faster RCNN-based curve text detector · Total-Text | F: 67.5 -> 80.2 | Helped |
| Use conditional spatial expansion instead of Mask R-CNN segmentation | Faster RCNN-based curve text detector · CTW-1500 | F: 67.8 -> 77.4 | Helped |
| Use conditional spatial expansion instead of polygon regression | curve text detector with shared backbone and RPN · Total-Text | F: 75.6 -> 80.6 | Helped |
| Use conditional spatial expansion instead of polygon regression | curve text detector with shared backbone and RPN · CTW-1500 | F: 73.2 -> 77.6 | Helped |
| Use CSE rather than the strongest listed peer method | curve text detection · Total-Text | F: 78.4 -> 80.2 | Helped |
| Use CSE rather than the strongest listed peer method | curve text detection · CTW-1500 | F: 75.6 -> 78.4 | Helped |
| Rescale the sampling grid by δ_s from 1.0 to 1.5 (ablation) | CSE · Total-Text | F: around 80% | No clear effect |
| Rescale the sampling grid by δ_s from 1.0 to 1.5 (ablation) | CSE · CTW-1500 | F: 78% | No clear effect |
| Increase sampling-grid rescaling beyond δ_s 1.5 (ablation) | CSE · Total-Text | F: above 77% | Hurt |
| Increase sampling-grid rescaling beyond δ_s 1.5 (ablation) | CSE · CTW-1500 | F: above 73% | Hurt |
| Relocate the seed within the target text, changing δ_c from 0.0 to 1.0 (ablation) | CSE · Total-Text | F: decrease by only 3% | Hurt |
| Relocate the seed within the target text, changing δ_c from 0.0 to 1.0 (ablation) | CSE · CTW-1500 | F: decrease by only 4% | Hurt |
Closest papers by meaning
- ContourNet: Taking a Further Step Toward Accurate Arbitrary-Shaped Scene Text Detection (CVPR 2020 · similarity 0.89)
- LRANet: Towards Accurate and Efficient Scene Text Detection with Low-Rank Approximation Network (AAAI 2024 · similarity 0.85)
- LayoutFormer: Hierarchical Text Detection Towards Scene Text Understanding (CVPR 2024 · similarity 0.84)
- Kernel Adaptive Convolution for Scene Text Detection via Distance Map Prediction (CVPR 2024 · similarity 0.84)
- Towards Unified Multi-granularity Text Detection with Interactive Attention (ICML 2024 · similarity 0.84)
- SRFormer: Text Detection Transformer with Incorporated Segmentation and Regression (AAAI 2024 · similarity 0.84)
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.