Research map / Detection and segmentation / CNN features
PGNet: Real-time Arbitrarily-Shaped Text Spotting with Point Gathering Network
Read the paperCheck 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 TDO to recover character reading order (ablation) | PGNet-A · ICDAR 2015 | Recognition accuracy: 61.9 -> 62.3; +0.4 | Helped |
| Add TDO to recover character reading order (ablation) | PGNet-E · ICDAR 2015 | Recognition accuracy: 56.9 -> 57.4; +0.5 | Helped |
| Add TDO to recover character reading order (ablation) | PGNet-A · Total-Text | Recognition accuracy: 60.4 -> 61.7; +1.3 | Helped |
| Add TDO to recover character reading order (ablation) | PGNet-E · Total-Text | Recognition accuracy: 56.8 -> 58.4; +1.6 | Helped |
| Add GRM for word-level semantic and visual context reasoning (ablation) | PGNet-A · ICDAR 2015 | Recognition accuracy: 62.3 -> 63.5; +1.2 | Helped |
| Add GRM for word-level semantic and visual context reasoning (ablation) | PGNet-E · ICDAR 2015 | Recognition accuracy: 57.4 -> 58.7; +1.3 | Helped |
| Add GRM for word-level semantic and visual context reasoning (ablation) | PGNet-A · Total-Text | Recognition accuracy: 61.7 -> 63.1; +1.4 | Helped |
| Add GRM for word-level semantic and visual context reasoning (ablation) | PGNet-E · Total-Text | Recognition accuracy: 58.4 -> 60.6; +2.2 | Helped |
Closest papers by meaning
- Text Perceptron: Towards End-to-End Arbitrary-Shaped Text Spotting (AAAI 2020 · similarity 0.87)
- LRANet: Towards Accurate and Efficient Scene Text Detection with Low-Rank Approximation Network (AAAI 2024 · similarity 0.85)
- Bridging the Gap Between End-to-End and Two-Step Text Spotting (CVPR 2024 · similarity 0.85)
- Arbitrary Reading Order Scene Text Spotter with Local Semantics Guidance (AAAI 2025 · similarity 0.84)
- SemiETS: Integrating Spatial and Content Consistencies for Semi-Supervised End-to-end Text Spotting (CVPR 2025 · similarity 0.84)
- GoMatching: A Simple Baseline for Video Text Spotting via Long and Short Term Matching (NeurIPS 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.