Research map / Detection and segmentation / CNN features
AE TextSpotter: Learning Visual and Linguistic Representation for Ambiguous Text Spotting
Wenhai Wang, Xuebo Liu, Xiaozhong Ji, Enze Xie, Ding Liang, ZhiBo Yang, Tong Lu, Chunhua Shen, Ping Luo
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 the language module to re-score candidate text lines (ablation) | AE TextSpotter · TDA-ReCTS | F-measure: 77.93 -> 81.39 | Improved |
| add the language module to re-score candidate text lines (ablation) | AE TextSpotter · TDA-ReCTS | 1-NED: 47.74 -> 51.32 | Improved |
| add the language module to re-score candidate text lines (ablation) | AE TextSpotter · IC19-ReCTS test | F-measure: 91.16 -> 91.80 | Improved |
| add the language module to re-score candidate text lines (ablation) | AE TextSpotter · IC19-ReCTS test | 1-NED: 70.95 -> 71.81 | Improved |
| add the CRM-based recognition and visual-only scoring on top of detection only (ablation) | AE TextSpotter w/o LM · TDA-ReCTS | F-measure: 77.19 -> 77.93 | Improved |
| enlarge input short side from 800 to 1000 | AE TextSpotter · IC19-ReCTS test | F-measure: 91.80 -> 91.86 | Improved |
| enlarge input short side from 800 to 1000 | AE TextSpotter · IC19-ReCTS test | 1-NED: 71.81 -> 73.12 | Improved |
| use loose TDM thresholds (score 0.01, NMS 0.9) instead of common ones (0.5, 0.5) (ablation) | Text detection module · TDA-ReCTS | recall: 0.852 -> 0.969 | Improved |
| loosen thresholds from (0.01, 0.9) to (0.0, 1.0) (ablation) | Text detection module · TDA-ReCTS | recall: 0.969 -> 0.971 | No clear effect |
| replace CRNN with the character-based recognition module (ablation) | AE TextSpotter recognition · IC19-ReCTS test | recognition time per image (ms): 966.7 -> 122.3 | Improved |
| replace ASTER with the character-based recognition module (ablation) | AE TextSpotter recognition · IC19-ReCTS test | recognition time per image (ms): 4828.2 -> 122.3 | Improved |
Later papers that compared with this method
1 later accepted paper reports AE TextSpotter as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- SwinTextSpotter: Scene Text Spotting via Better Synergy Between Text Detection and Text Recognition (CVPR 2022 · better on 1 of 1 shared result)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- FOTS — FOTS: Fast Oriented Text Spotting With a Unified Network (CVPR 2018) · better on 4 of 4 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- SwinTextSpotter: Scene Text Spotting via Better Synergy Between Text Detection and Text Recognition (CVPR 2022 · similarity 0.89)
- SemiETS: Integrating Spatial and Content Consistencies for Semi-Supervised End-to-end Text Spotting (CVPR 2025 · similarity 0.87)
- Bridging the Gap Between End-to-End and Two-Step Text Spotting (CVPR 2024 · similarity 0.87)
- GoMatching: A Simple Baseline for Video Text Spotting via Long and Short Term Matching (NeurIPS 2024 · similarity 0.87)
- Arbitrary Reading Order Scene Text Spotter with Local Semantics Guidance (AAAI 2025 · similarity 0.87)
- OTE: Exploring Accurate Scene Text Recognition Using One Token (CVPR 2024 · 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.
