Research map / Detection and segmentation / CNN features
Text Spotting Transformers
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 bounding-box guidance to the control point queries (ablation) | TESTR-Polygon · Total-Text | End-to-end: 67.5 -> 73.3 | Helped |
| use multi-scale features instead of only the last-stage ResNet feature (ablation) | TESTR-Polygon · Total-Text | End-to-end: 62.5 -> 73.3 | Helped |
| increase polygon input scale from 720 to 1600 (ablation) | TESTR-Polygon · Total-Text | End-to-end: 66.2 -> 73.3 | Helped |
| increase polygon input scale from 720 to 1600 (ablation) | TESTR-Polygon · Total-Text | FPS: 11.7 -> 5.3 | Hurt |
| set lambda_char to 8.0 rather than 4.0 (ablation) | TESTR-Polygon · Total-Text | End-to-end F: 73.3 -> 73.8 | Helped |
| use polygon annotations rather than Bezier annotations | TESTR · Total-Text | End-to-end None: 71.6 -> 73.3 | Helped |
| use polygon annotations rather than Bezier annotations | TESTR · CTW1500 | End-to-end None: 53.3 -> 56.0 | Mixed |
Closest papers by meaning
- Bridging the Gap Between End-to-End and Two-Step Text Spotting (CVPR 2024 · similarity 0.87)
- Box2Poly: Memory-Efficient Polygon Prediction of Arbitrarily Shaped and Rotated Text (AAAI 2024 · similarity 0.87)
- GoMatching: A Simple Baseline for Video Text Spotting via Long and Short Term Matching (NeurIPS 2024 · similarity 0.86)
- SemiETS: Integrating Spatial and Content Consistencies for Semi-Supervised End-to-end Text Spotting (CVPR 2025 · similarity 0.85)
- General Detection-based Text Line Recognition (NeurIPS 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.