Optical Character Recognition2020en
scut-ctw1500
Dataset from Papers With Code
Metrics:accuracy, cer, wer, f1
f-measure
| # | Model | Score | Paper / Code | Date |
|---|---|---|---|---|
| 1 | MixNet | 89.8 | Aug 2023 | |
| 2 | SRFormer (ResNet-50) | 89.6 | Aug 2023 | |
| 3 | DPText-DETR (ResNet50) | 88.8 | Jul 2022 | |
| 4 | TextFuseNet (ResNeXt-101) | 87.4 | TextFuseNet: Scene Text Detection with Richer Fused FeaturesCode | May 2020 |
| 5 | I3CL + SSL | 86.5 | Aug 2021 | |
| 6 | PAN | 85 | Nov 2018 | |
| 7 | FAST-B-640 | 84.2 | Nov 2021 | |
| 8 | PAN-640 | 83.7 | Aug 2019 | |
| 9 | CRAFT | 83.5 | Apr 2019 | |
| 10 | DB-ResNet50 (1024) | 83.4 | Nov 2019 | |
| 11 | FAST-B-512 | 82.9 | Nov 2021 | |
| 12 | FAST-S-512 | 82 | Nov 2021 | |
| 13 | FAST-T-512 | 81.5 | Nov 2021 | |
| 14 | PSENet-1s | 81.17 | Jun 2018 | |
| 15 | TextSnake | 75.6 | Jul 2018 | |
| 16 | SLPR | 74.8 | Jan 2018 |
f-measure-full-lexicon
| # | Model | Score | Paper / Code | Date |
|---|---|---|---|---|
| 1 | SPTS | 83.8 | Dec 2021 | |
| 2 | A3S | 82.3 | Feb 2023 | |
| 3 | TESTR | 81.5 | Apr 2022 | |
| 4 | DeepSolo (ResNet-50) | 81.4 | May 2023 | |
| 5 | ABINet++ | 80.3 | Nov 2022 | |
| 6 | TPSNet | 79.2 | Oct 2021 | |
| 7 | MANGO | 78.7 | Dec 2020 | |
| 8 | ABCNet v2 | 77.2 | May 2021 | |
| 9 | SwinTextSpotter | 77 | Mar 2022 | |
| 10 | TextDragon | 72.4 | TextDragon: An End-to-End Framework for Arbitrary Shaped Text Spotting | Oct 2019 |
f-measure-no-lexicon
| # | Model | Score | Paper / Code | Date |
|---|---|---|---|---|
| 1 | A3S | 64.4 | Feb 2023 | |
| 2 | DeepSolo (ResNet-50) | 64.2 | May 2023 | |
| 3 | SPTS | 63.6 | Dec 2021 | |
| 4 | ABINet++ | 60.2 | Nov 2022 | |
| 5 | TPSNet | 59.7 | Oct 2021 | |
| 6 | MANGO | 58.9 | Dec 2020 | |
| 7 | ABCNet v2 | 57.5 | May 2021 | |
| 8 | TextPerceptron | 57 | Feb 2020 | |
| 9 | TESTR | 56 | Apr 2022 | |
| 10 | SwinTextSpotter | 51.8 | Mar 2022 | |
| 11 | TextDragon | 39.7 | TextDragon: An End-to-End Framework for Arbitrary Shaped Text Spotting | Oct 2019 |
fps
| # | Model | Score | Paper / Code | Date |
|---|---|---|---|---|
| 1 | FAST-T-512 | 129.1 | Nov 2021 | |
| 2 | FAST-S-512 | 112.9 | Nov 2021 | |
| 3 | FAST-B-512 | 92.6 | Nov 2021 | |
| 4 | FAST-B-640 | 66.5 | Nov 2021 | |
| 5 | PAN | 65.2 | Nov 2018 | |
| 6 | MixNet | 15.2 | Aug 2023 |
precision
| # | Model | Score | Paper / Code | Date |
|---|---|---|---|---|
| 1 | DPText-DETR (ResNet50) | 91.7 | Jul 2022 | |
| 2 | SRFormer (ResNet-50) | 91.6 | Aug 2023 | |
| 3 | MixNet | 91.4 | Aug 2023 | |
| 4 | TextFuseNet (ResNeXt-101) | 89.7 | TextFuseNet: Scene Text Detection with Richer Fused FeaturesCode | May 2020 |
| 5 | I3CL + SSL | 88.4 | Aug 2021 | |
| 6 | FAST-B-640 | 87.8 | Nov 2021 | |
| 7 | PAN | 86.8 | Nov 2018 | |
| 8 | PAN-640 | 86.4 | Aug 2019 | |
| 9 | CRAFT | 86 | Apr 2019 | |
| 10 | FAST-B-512 | 85.7 | Nov 2021 | |
| 11 | FAST-S-512 | 85.6 | Nov 2021 | |
| 12 | FAST-T-512 | 85.5 | Nov 2021 | |
| 13 | PSENet-1s | 82.5 | Jun 2018 | |
| 14 | SLPR | 80.1 | Jan 2018 | |
| 15 | TextSnake | 67.9 | Jul 2018 |
recall
| # | Model | Score | Paper / Code | Date |
|---|---|---|---|---|
| 1 | MixNet | 88.3 | Aug 2023 | |
| 2 | SRFormer (ResNet-50) | 87.7 | Aug 2023 | |
| 3 | DPText-DETR (ResNet50) | 86.2 | Jul 2022 | |
| 4 | TextSnake | 85.3 | Jul 2018 | |
| 5 | TextFuseNet (ResNeXt-101) | 85.1 | TextFuseNet: Scene Text Detection with Richer Fused FeaturesCode | May 2020 |
| 6 | I3CL + SSL | 84.6 | Aug 2021 | |
| 7 | PAN | 83.2 | Nov 2018 | |
| 8 | PAN-640 | 81.2 | Aug 2019 | |
| 9 | CRAFT | 81.1 | Apr 2019 | |
| 10 | FAST-B-640 | 80.9 | Nov 2021 | |
| 11 | FAST-B-512 | 80.2 | Nov 2021 | |
| 12 | PSENet-1s | 79.89 | Jun 2018 | |
| 13 | FAST-S-512 | 78.7 | Nov 2021 | |
| 14 | FAST-T-512 | 77.9 | Nov 2021 | |
| 15 | SLPR | 70.1 | Jan 2018 |
Related Papers22
MixNet: Toward Accurate Detection of Challenging Scene Text in the Wild
Aug 2023Models: MixNet
SRFormer: Text Detection Transformer with Incorporated Segmentation and Regression
Aug 2023Models: SRFormer (ResNet-50)
DeepSolo++: Let Transformer Decoder with Explicit Points Solo for Multilingual Text Spotting
May 2023Models: DeepSolo (ResNet-50)
DPText-DETR: Towards Better Scene Text Detection with Dynamic Points in Transformer
Jul 2022Models: DPText-DETR (ResNet50)
Text Spotting Transformers
Apr 2022Models: TESTR
SwinTextSpotter: Scene Text Spotting via Better Synergy between Text Detection and Text Recognition
Mar 2022Models: SwinTextSpotter
SPTS: Single-Point Text Spotting
Dec 2021Models: SPTS
FAST: Faster Arbitrarily-Shaped Text Detector with Minimalist Kernel Representation
Nov 2021Models: FAST-B-640, FAST-B-512, FAST-S-512 +1 more
I3CL:Intra- and Inter-Instance Collaborative Learning for Arbitrary-shaped Scene Text Detection
Aug 2021Models: I3CL + SSL
ABCNet v2: Adaptive Bezier-Curve Network for Real-time End-to-end Text Spotting
May 2021Models: ABCNet v2
MANGO: A Mask Attention Guided One-Stage Scene Text Spotter
Dec 2020Models: MANGO
Text Perceptron: Towards End-to-End Arbitrary-Shaped Text Spotting
Feb 2020Models: TextPerceptron
Real-time Scene Text Detection with Differentiable Binarization
Nov 2019Models: DB-ResNet50 (1024)
Character Region Awareness for Text Detection
Apr 2019Models: CRAFT
Mask R-CNN with Pyramid Attention Network for Scene Text Detection
Nov 2018Models: PAN
TextSnake: A Flexible Representation for Detecting Text of Arbitrary Shapes
Jul 2018Models: TextSnake
Shape Robust Text Detection with Progressive Scale Expansion Network
Jun 2018Models: PSENet-1s
Sliding Line Point Regression for Shape Robust Scene Text Detection
Jan 2018Models: SLPR