Scene Text Detection2020en
coco-text
Dataset from Papers With Code
Metrics:accuracy, cer, wer, f1
1-1-accuracy
| # | Model | Score | Paper / Code | Date |
|---|---|---|---|---|
| 1 | CLIP4STR-L | 81.9 | May 2023 | |
| 2 | MGP-STR | 81.7 | Sep 2022 | |
| 3 | CLIP4STR-B* | 81.1 | May 2023 |
f-measure
| # | Model | Score | Paper / Code | Date |
|---|---|---|---|---|
| 1 | Corner-based Region Proposals | 59.1 | Apr 2018 | |
| 2 | TextBoxes++_MS | 58.72 | Jan 2018 | |
| 3 | EAST + VGG16 | 39.45 | Apr 2017 | |
| 4 | SSTD | 37 | Sep 2017 | |
| 5 | WordSup (VGG16-synth-coco) | 36.8 | Aug 2017 | |
| 6 | Yao et al. | 33.31 | Jun 2016 |
precision
| # | Model | Score | Paper / Code | Date |
|---|---|---|---|---|
| 1 | TextBoxes++_MS | 60.87 | Jan 2018 | |
| 2 | Corner-based Region Proposals | 55.5 | Apr 2018 | |
| 3 | EAST + VGG16 | 50.39 | Apr 2017 | |
| 4 | SSTD | 46 | Sep 2017 | |
| 5 | WordSup (VGG16-synth-coco) | 45.2 | Aug 2017 | |
| 6 | Yao et al. | 43.23 | Jun 2016 |
recall
| # | Model | Score | Paper / Code | Date |
|---|---|---|---|---|
| 1 | Corner-based Region Proposals | 63.3 | Apr 2018 | |
| 2 | TextBoxes++_MS | 56.7 | Jan 2018 | |
| 3 | EAST + VGG16 | 32.4 | Apr 2017 | |
| 4 | SSTD | 31 | Sep 2017 | |
| 5 | WordSup (VGG16-synth-coco) | 30.9 | Aug 2017 | |
| 6 | Yao et al. | 27.1 | Jun 2016 |
Related Papers8
CLIP4STR: A Simple Baseline for Scene Text Recognition with Pre-trained Vision-Language Model
May 2023Models: CLIP4STR-L, CLIP4STR-B*
Multi-Granularity Prediction for Scene Text Recognition
Sep 2022Models: MGP-STR
Detecting Multi-Oriented Text with Corner-based Region Proposals
Apr 2018Models: Corner-based Region Proposals
TextBoxes++: A Single-Shot Oriented Scene Text Detector
Jan 2018Models: TextBoxes++_MS
Single Shot Text Detector with Regional Attention
Sep 2017Models: SSTD
WordSup: Exploiting Word Annotations for Character based Text Detection
Aug 2017Models: WordSup (VGG16-synth-coco)
EAST: An Efficient and Accurate Scene Text Detector
Apr 2017Models: EAST + VGG16
Scene Text Detection via Holistic, Multi-Channel Prediction
Jun 2016Models: Yao et al.