Research map / Label-efficient and robust learning / Recognition and long-tail
Implicit Feature Alignment: Learn To Convert Text Recognizer to Text Spotter
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 WH-ACE (ablation) | ADP with IFA-inference · IAM | CER: 27.65 -> 10.28 | Helped |
| add WH-ACE (ablation) | ExCTC with IFA-inference · IAM | CER: 15.83 -> 6.26 | Helped |
| add WH-ACE (ablation) | ADP with conventional inference · IAM | CER: 11.85 -> 11.78 | Helped |
| add WH-ACE (ablation) | ExCTC with conventional inference · IAM | CER: 6.48 -> 6.42 | Helped |
| replace conventional decoding with IFA-inference (ablation) | ADP · IAM | CER: 11.78 -> 10.28 | Helped |
| replace conventional decoding with IFA-inference (ablation) | ExCTC · IAM | CER: 6.42 -> 6.26 | Helped |
| apply IFA-inference to full-page rather than line input (ablation) | ExCTC · IAM | CER: 6.26 -> 6.16 | Helped |
| apply IFA-inference to full-page rather than line input (ablation) | ADP · IAM | CER: 10.28 -> 9.97 | Helped |
| use ExCTC instead of ADP | IFA full-page recognition · IAM | CER: 9.97 -> 6.16 | Helped |
| use ADP instead of attention-based WAP + TD | OffRaSHME character recognition · OffRaSHME | F-measure: 84.74 -> 89.32 | Helped |
| use ExCTC instead of Det + Recog | detection-free text spotting · ICDAR-2013 | Speed: 2.79 -> 31.76 | Helped |
| use IFA rather than Det + Recog | Touching-Block · Touching-Block | AR: 77.75 -> 84.96 | Helped |
Closest papers by meaning
- ODM: A Text-Image Further Alignment Pre-training Approach for Scene Text Detection and Spotting (CVPR 2024 · similarity 0.86)
- General Detection-based Text Line Recognition (NeurIPS 2024 · similarity 0.85)
- OCRBench v2: An Improved Benchmark for Evaluating Large Multimodal Models on Visual Text Localization and Reasoning (NeurIPS 2025 · similarity 0.85)
- Bridging the Gap Between End-to-End and Two-Step Text Spotting (CVPR 2024 · similarity 0.85)
- OTE: Exploring Accurate Scene Text Recognition Using One Token (CVPR 2024 · similarity 0.85)
- Out of Length Text Recognition with Sub-String Matching (AAAI 2025 · 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.