Research map / Detection and segmentation / CNN features
Look More Than Once: An Accurate Detector for Text of Arbitrary Shapes
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 IRM with one refinement pass (ablation) | DR · ICDAR2017-RCTW | Hmean: 58.96 -> 62.30 | Helped |
| increase IRM refinement passes from 1 to 2 (ablation) | DR+IRM · ICDAR2017-RCTW | Hmean: 62.30 -> 62.62 | Helped |
| increase IRM refinement passes from 2 to 3 (ablation) | DR+IRM · ICDAR2017-RCTW | Hmean: 62.62 -> 62.73 | Helped |
| add corner attention maps (ablation) | DR+IRM · ICDAR2017-RCTW | Hmean IoU@0.5: 62.31 -> 62.62 | Helped |
| add corner attention maps (ablation) | DR+IRM · ICDAR2017-RCTW | Hmean IoU@0.7: 51.64 -> 53.14 | Helped |
| add SEM (ablation) | DR · SCUT-CTW1500 | Hmean: 70.58 -> 77.75 | Helped |
| add SEM (ablation) | DR+IRM · SCUT-CTW1500 | Hmean: 72.09 -> 78.43 | Helped |
| use multi-scale testing | LOMO · SCUT-CTW1500 | Hmean: 78.4 -> 80.8 | Helped |
| use multi-scale testing | LOMO · Total-Text | Hmean: 81.6 -> 83.3 | Helped |
| use multi-scale testing | LOMO · ICDAR 2015 | Hmean: 87.2 -> 87.7 | Helped |
Closest papers by meaning
- Arbitrary Shape Scene Text Detection With Adaptive Text Region Representation (CVPR 2019 · similarity 0.90)
- LRANet: Towards Accurate and Efficient Scene Text Detection with Low-Rank Approximation Network (AAAI 2024 · similarity 0.89)
- Text Grouping Adapter: Adapting Pre-trained Text Detector for Layout Analysis (CVPR 2024 · similarity 0.87)
- Towards Unified Multi-granularity Text Detection with Interactive Attention (ICML 2024 · similarity 0.87)
- Kernel Adaptive Convolution for Scene Text Detection via Distance Map Prediction (CVPR 2024 · similarity 0.87)
- OCRBench v2: An Improved Benchmark for Evaluating Large Multimodal Models on Visual Text Localization and Reasoning (NeurIPS 2025 · similarity 0.86)
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.