未验证 提交 4725a557 编写于 作者: Z zhoujun 提交者: GitHub

add can, stt to algorithm_overview_en.md (#8317)

上级 58a5c0b3
...@@ -3,6 +3,8 @@ ...@@ -3,6 +3,8 @@
- [1. 两阶段OCR算法](#1) - [1. 两阶段OCR算法](#1)
- [1.1 文本检测算法](#11) - [1.1 文本检测算法](#11)
- [1.2 文本识别算法](#12) - [1.2 文本识别算法](#12)
- [1.3 文本超分辨率算法](#13)
- [1.4 公式识别算法](#14)
- [2. 端到端OCR算法](#2) - [2. 端到端OCR算法](#2)
- [3. 表格识别算法](#3) - [3. 表格识别算法](#3)
- [4. 关键信息抽取算法](#4) - [4. 关键信息抽取算法](#4)
...@@ -107,6 +109,34 @@ PaddleOCR将**持续新增**支持OCR领域前沿算法与模型,**欢迎广 ...@@ -107,6 +109,34 @@ PaddleOCR将**持续新增**支持OCR领域前沿算法与模型,**欢迎广
|RobustScanner|ResNet31| 87.77% | rec_r31_robustscanner | [训练模型](https://paddleocr.bj.bcebos.com/contribution/rec_r31_robustscanner.tar)| |RobustScanner|ResNet31| 87.77% | rec_r31_robustscanner | [训练模型](https://paddleocr.bj.bcebos.com/contribution/rec_r31_robustscanner.tar)|
|RFL|ResNetRFL| 88.63% | rec_resnet_rfl_att | [训练模型](https://paddleocr.bj.bcebos.com/contribution/rec_resnet_rfl_att_train.tar) | |RFL|ResNetRFL| 88.63% | rec_resnet_rfl_att | [训练模型](https://paddleocr.bj.bcebos.com/contribution/rec_resnet_rfl_att_train.tar) |
<a name="13"></a>
### 1.3 文本超分辨率算法
已支持的文本超分辨率算法列表(戳链接获取使用教程):
- [x] [Text Gestalt](./algorithm_sr_gestalt.md)
- [x] [Text Telescope](./algorithm_sr_telescope.md)
在TextZoom公开数据集上,算法效果如下:
|模型|骨干网络|PSNR_Avg|SSIM_Avg|配置文件|下载链接|
|---|---|---|---|---|---|
|Text Gestalt|tsrn|19.28|0.6560| [configs/sr/sr_tsrn_transformer_strock.yml](../../configs/sr/sr_tsrn_transformer_strock.yml)|[训练模型](https://paddleocr.bj.bcebos.com/sr_tsrn_transformer_strock_train.tar)|
|Text Telescope|tbsrn|21.56|0.7411| [configs/sr/sr_telescope.yml](../../configs/sr/sr_telescope.yml)|[训练模型](https://paddleocr.bj.bcebos.com/contribution/sr_telescope_train.tar)|
<a name="14"></a>
### 1.4 公式识别算法
已支持的公式识别算法列表(戳链接获取使用教程):
- [x] [CAN](./algorithm_rec_can.md.md)
在CROHME手写公式数据集上,算法效果如下:
|模型 |骨干网络|配置文件|ExpRate|下载链接|
| ----- | ----- | ----- | ----- | ----- |
|CAN|DenseNet|[rec_d28_can.yml](../../configs/rec/rec_d28_can.yml)|51.72%|[训练模型](https://paddleocr.bj.bcebos.com/contribution/rec_d28_can_train.tar)|
<a name="2"></a> <a name="2"></a>
## 2. 端到端算法 ## 2. 端到端算法
......
...@@ -3,6 +3,8 @@ ...@@ -3,6 +3,8 @@
- [1. Two-stage OCR Algorithms](#1) - [1. Two-stage OCR Algorithms](#1)
- [1.1 Text Detection Algorithms](#11) - [1.1 Text Detection Algorithms](#11)
- [1.2 Text Recognition Algorithms](#12) - [1.2 Text Recognition Algorithms](#12)
- [1.3 Text Super-Resolution Algorithms](#13)
- [1.4 Formula Recognition Algorithm](#14)
- [2. End-to-end OCR Algorithms](#2) - [2. End-to-end OCR Algorithms](#2)
- [3. Table Recognition Algorithms](#3) - [3. Table Recognition Algorithms](#3)
- [4. Key Information Extraction Algorithms](#4) - [4. Key Information Extraction Algorithms](#4)
...@@ -104,6 +106,36 @@ Refer to [DTRB](https://arxiv.org/abs/1904.01906), the training and evaluation r ...@@ -104,6 +106,36 @@ Refer to [DTRB](https://arxiv.org/abs/1904.01906), the training and evaluation r
|RobustScanner|ResNet31| 87.77% | rec_r31_robustscanner | [trained model](https://paddleocr.bj.bcebos.com/contribution/rec_r31_robustscanner.tar)| |RobustScanner|ResNet31| 87.77% | rec_r31_robustscanner | [trained model](https://paddleocr.bj.bcebos.com/contribution/rec_r31_robustscanner.tar)|
|RFL|ResNetRFL| 88.63% | rec_resnet_rfl_att | [trained model](https://paddleocr.bj.bcebos.com/contribution/rec_resnet_rfl_att_train.tar) | |RFL|ResNetRFL| 88.63% | rec_resnet_rfl_att | [trained model](https://paddleocr.bj.bcebos.com/contribution/rec_resnet_rfl_att_train.tar) |
<a name="13"></a>
### 1.3 Text Super-Resolution Algorithms
Supported text super-resolution algorithms (Click the link to get the tutorial):
- [x] [Text Gestalt](./algorithm_sr_gestalt.md)
- [x] [Text Telescope](./algorithm_sr_telescope.md)
On the TextZoom public dataset, the effect of the algorithm is as follows:
|Model|Backbone|PSNR_Avg|SSIM_Avg|Config|Download link|
|---|---|---|---|---|---|
|Text Gestalt|tsrn|19.28|0.6560| [configs/sr/sr_tsrn_transformer_strock.yml](../../configs/sr/sr_tsrn_transformer_strock.yml)|[trained model](https://paddleocr.bj.bcebos.com/sr_tsrn_transformer_strock_train.tar)|
|Text Telescope|tbsrn|21.56|0.7411| [configs/sr/sr_telescope.yml](../../configs/sr/sr_telescope.yml)|[trained model](https://paddleocr.bj.bcebos.com/contribution/sr_telescope_train.tar)|
<a name="14"></a>
### 1.4 Formula Recognition Algorithm
Supported formula recognition algorithms (Click the link to get the tutorial):
- [x] [CAN](./algorithm_rec_can.md.md)
On the CROHME handwritten formula dataset, the effect of the algorithm is as follows:
|Model |Backbone|Config|ExpRate|Download link|
| ----- | ----- | ----- | ----- | ----- |
|CAN|DenseNet|[rec_d28_can.yml](../../configs/rec/rec_d28_can.yml)|51.72%|[trained model](https://paddleocr.bj.bcebos.com/contribution/rec_d28_can_train.tar)|
<a name="2"></a> <a name="2"></a>
## 2. End-to-end OCR Algorithms ## 2. End-to-end OCR Algorithms
...@@ -122,7 +154,7 @@ On the PubTabNet dataset, the algorithm result is as follows: ...@@ -122,7 +154,7 @@ On the PubTabNet dataset, the algorithm result is as follows:
|Model|Backbone|Config|Acc|Download link| |Model|Backbone|Config|Acc|Download link|
|---|---|---|---|---| |---|---|---|---|---|
|TableMaster|TableResNetExtra|[configs/table/table_master.yml](../../configs/table/table_master.yml)|77.47%|[trained](https://paddleocr.bj.bcebos.com/ppstructure/models/tablemaster/table_structure_tablemaster_train.tar) / [inference model](https://paddleocr.bj.bcebos.com/ppstructure/models/tablemaster/table_structure_tablemaster_infer.tar)| |TableMaster|TableResNetExtra|[configs/table/table_master.yml](../../configs/table/table_master.yml)|77.47%|[trained model](https://paddleocr.bj.bcebos.com/ppstructure/models/tablemaster/table_structure_tablemaster_train.tar) / [inference model](https://paddleocr.bj.bcebos.com/ppstructure/models/tablemaster/table_structure_tablemaster_infer.tar)|
<a name="4"></a> <a name="4"></a>
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册