| MixNet: Toward Accurate Detection of Challenging Scene Text in the Wild | ✓ Link | 90.5% | 93.0 | 88.1 | 15.2 | MixNet | 2023-08-23 |
| SRFormer: Text Detection Transformer with Incorporated Segmentation and Regression | ✓ Link | 90.0% | 92.2% | 87.9% | | SRFormer (ResNet-50) | 2023-08-21 |
| DPText-DETR: Towards Better Scene Text Detection with Dynamic Points in Transformer | ✓ Link | 89.0% | 91.8% | 86.4% | 17 | DPText-DETR (ResNet-50) | 2022-07-10 |
| FAST: Faster Arbitrarily-Shaped Text Detector with Minimalist Kernel Representation | ✓ Link | 87.5% | 90.0 | 85.2 | 46 | FAST-B-800 | 2021-11-03 |
| TextFuseNet: Scene Text Detection with Richer Fused Features | ✓ Link | 87.5% | 89.2 | 85.8 | | TextFuseNet (ResNeXt-101) | 2020-05-17 |
| I3CL:Intra- and Inter-Instance Collaborative Learning for Arbitrary-shaped Scene Text Detection | ✓ Link | 86.9% | 89.8 | 84.2 | | I3CL + SSL(ResNet-50) | 2021-08-03 |
| Convolutional Character Networks | ✓ Link | 86.5% | 88 | 85 | | CharNet H-88 (multi-scale) | 2019-10-17 |
| FAST: Faster Arbitrarily-Shaped Text Detector with Minimalist Kernel Representation | ✓ Link | 86.4% | 89.9 | 83.2 | 67.5 | FAST-B-640 | 2021-11-03 |
| Real-Time Scene Text Detection with Differentiable Binarization and Adaptive Scale Fusion | ✓ Link | 86% | 88.9 | 83.2 | 28 | DBNet++ (ResNet-50) (800) | 2022-02-21 |
| FAST: Faster Arbitrarily-Shaped Text Detector with Minimalist Kernel Representation | ✓ Link | 85.8% | 89.6 | 82.4 | 93.2 | FAST-B-512 | 2021-11-03 |
| Convolutional Character Networks | ✓ Link | 85.6% | 89.9 | 81.7 | | CharNet H-88 | 2019-10-17 |
| A method for detecting text of arbitrary shapes in natural scenes that improves text spotting | | 85.6% | | | | SA-Text | 2019-11-16 |
| Efficient and Accurate Arbitrary-Shaped Text Detection with Pixel Aggregation Network | ✓ Link | 85% | 89.3 | 81 | | PAN-640 | 2019-08-16 |
| FAST: Faster Arbitrarily-Shaped Text Detector with Minimalist Kernel Representation | ✓ Link | 84.9% | 88.3 | 81.7 | 115.5 | FAST-S-512 | 2021-11-03 |
| Real-time Scene Text Detection with Differentiable Binarization | ✓ Link | 84.7% | | | | DB-ResNet-50 (800) | 2019-11-20 |
| TextCohesion: Detecting Text for Arbitrary Shapes | | 84.6% | | | | TextCohesion | 2019-04-22 |
| Character Region Awareness for Text Detection | ✓ Link | 83.6% | 87.6 | 79.9 | | CRAFT | 2019-04-03 |
| Real-Time Scene Text Detection with Differentiable Binarization and Adaptive Scale Fusion | ✓ Link | 83.3% | 87.4 | 79.6 | 48 | DBNet++ (ResNet-18) (800) | 2022-02-21 |
| Scene Text Detection with Supervised Pyramid Context Network | ✓ Link | 82.9% | 83 | 82.8 | | SPCNET | 2018-11-21 |
| FAST: Faster Arbitrarily-Shaped Text Detector with Minimalist Kernel Representation | ✓ Link | 81.6% | 86.5 | 77.2 | 152.8 | FAST-T-448 | 2021-11-03 |
| Fused Text Segmentation Networks for Multi-oriented Scene Text Detection | | 81.3% | 84.7 | 78 | | FTSN | 2017-09-11 |
| TextField: Learning A Deep Direction Field for Irregular Scene Text Detection | ✓ Link | 80.6% | 81.2 | 79.9 | | TextFiled | 2018-12-04 |
| Shape Robust Text Detection with Progressive Scale Expansion Network | ✓ Link | 79.6% | 84.5 | 75.2 | | PSENet-4s | 2019-03-28 |
| TextSnake: A Flexible Representation for Detecting Text of Arbitrary Shapes | ✓ Link | 78.4% | 82.7 | 74.5 | | TextSnake | 2018-07-04 |
| Mask TextSpotter: An End-to-End Trainable Neural Network for Spotting Text with Arbitrary Shapes | ✓ Link | 61.3% | 69 | 55 | | Mask TextSpotter | 2018-07-06 |
| EAST: An Efficient and Accurate Scene Text Detector | ✓ Link | 42.0% | 50.0 | 36.2 | | EAST | 2017-04-11 |
| Total-Text: A Comprehensive Dataset for Scene Text Detection and Recognition | ✓ Link | 36.0% | 40.0 | 33.0 | | Ch,ng et al. | 2017-10-28 |