README.md 7.3 KB
Newer Older
文幕地方's avatar
文幕地方 已提交
1 2 3 4 5 6 7 8 9 10
- [Table Recognition](#table-recognition)
  - [1. pipeline](#1-pipeline)
  - [2. Performance](#2-performance)
  - [3. How to use](#3-how-to-use)
    - [3.1 quick start](#31-quick-start)
    - [3.2 Train](#32-train)
    - [3.3 Eval](#33-eval)
    - [3.4 Inference](#34-inference)


M
MissPenguin 已提交
11
# Table Recognition
W
WenmuZhou 已提交
12 13

## 1. pipeline
M
MissPenguin 已提交
14
The table recognition mainly contains three models
W
WenmuZhou 已提交
15 16 17 18
1. Single line text detection-DB
2. Single line text recognition-CRNN
3. Table structure and cell coordinate prediction-RARE

M
MissPenguin 已提交
19
The table recognition flow chart is as follows
W
WenmuZhou 已提交
20

M
MissPenguin 已提交
21
![tableocr_pipeline](../docs/table/tableocr_pipeline_en.jpg)
W
WenmuZhou 已提交
22 23 24 25 26 27

1. The coordinates of single-line text is detected by DB model, and then sends it to the recognition model to get the recognition result.
2. The table structure and cell coordinates is predicted by RARE model.
3. The recognition result of the cell is combined by the coordinates, recognition result of the single line and the coordinates of the cell.
4. The cell recognition result and the table structure together construct the html string of the table.

W
WenmuZhou 已提交
28 29
## 2. Performance
We evaluated the algorithm on the PubTabNet<sup>[1]</sup> eval dataset, and the performance is as follows:
W
WenmuZhou 已提交
30

W
WenmuZhou 已提交
31 32

|Method|[TEDS(Tree-Edit-Distance-based Similarity)](https://github.com/ibm-aur-nlp/PubTabNet/tree/master/src)|
33 34 35
| --- | --- |
| EDD<sup>[2]</sup> | 88.3 |
| Ours | 93.32 |
W
WenmuZhou 已提交
36 37 38 39

## 3. How to use

### 3.1 quick start
W
WenmuZhou 已提交
40

W
WenmuZhou 已提交
41 42 43 44 45
```python
cd PaddleOCR/ppstructure

# download model
mkdir inference && cd inference
46 47 48 49 50
# Download the detection model of the ultra-lightweight table English OCR model and unzip it
wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/table/en_ppocr_mobile_v2.0_table_det_infer.tar && tar xf en_ppocr_mobile_v2.0_table_det_infer.tar
# Download the recognition model of the ultra-lightweight table English OCR model and unzip it
wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/table/en_ppocr_mobile_v2.0_table_rec_infer.tar && tar xf en_ppocr_mobile_v2.0_table_rec_infer.tar
# Download the ultra-lightweight English table inch model and unzip it
W
WenmuZhou 已提交
51 52
wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/table/en_ppocr_mobile_v2.0_table_structure_infer.tar && tar xf en_ppocr_mobile_v2.0_table_structure_infer.tar
cd ..
53
# run
54
python3 table/predict_table.py --det_model_dir=inference/en_ppocr_mobile_v2.0_table_det_infer --rec_model_dir=inference/en_ppocr_mobile_v2.0_table_rec_infer --table_model_dir=inference/en_ppocr_mobile_v2.0_table_structure_infer --image_dir=./docs/table/table.jpg --rec_char_dict_path=../ppocr/utils/dict/table_dict.txt --table_char_dict_path=../ppocr/utils/dict/table_structure_dict.txt --det_limit_side_len=736 --det_limit_type=min --output ./output/table
W
WenmuZhou 已提交
55
```
56 57
Note: The above model is trained on the PubLayNet dataset and only supports English scanning scenarios. If you need to identify other scenarios, you need to train the model yourself and replace the three fields `det_model_dir`, `rec_model_dir`, `table_model_dir`.

W
WenmuZhou 已提交
58 59
After running, the excel sheet of each picture will be saved in the directory specified by the output field

W
WenmuZhou 已提交
60
### 3.2 Train
W
WenmuZhou 已提交
61 62 63

In this chapter, we only introduce the training of the table structure model, For model training of [text detection](../../doc/doc_en/detection_en.md) and [text recognition](../../doc/doc_en/recognition_en.md), please refer to the corresponding documents

文幕地方's avatar
文幕地方 已提交
64
* data preparation  
W
WenmuZhou 已提交
65 66
The training data uses public data set [PubTabNet](https://arxiv.org/abs/1911.10683 ), Can be downloaded from the official [website](https://github.com/ibm-aur-nlp/PubTabNet) 。The PubTabNet data set contains about 500,000 images, as well as annotations in html format。

文幕地方's avatar
文幕地方 已提交
67
* Start training  
W
WenmuZhou 已提交
68 69 70 71 72 73 74 75 76 77 78 79
*If you are installing the cpu version of paddle, please modify the `use_gpu` field in the configuration file to false*
```shell
# single GPU training
python3 tools/train.py -c configs/table/table_mv3.yml
# multi-GPU training
# Set the GPU ID used by the '--gpus' parameter.
python3 -m paddle.distributed.launch --gpus '0,1,2,3' tools/train.py -c configs/table/table_mv3.yml
```

In the above instruction, use `-c` to select the training to use the `configs/table/table_mv3.yml` configuration file.
For a detailed explanation of the configuration file, please refer to [config](../../doc/doc_en/config_en.md).

文幕地方's avatar
文幕地方 已提交
80
* load trained model and continue training
W
WenmuZhou 已提交
81 82 83 84 85 86 87 88

If you expect to load trained model and continue the training again, you can specify the parameter `Global.checkpoints` as the model path to be loaded.

```shell
python3 tools/train.py -c configs/table/table_mv3.yml -o Global.checkpoints=./your/trained/model
```

**Note**: The priority of `Global.checkpoints` is higher than that of `Global.pretrain_weights`, that is, when two parameters are specified at the same time, the model specified by `Global.checkpoints` will be loaded first. If the model path specified by `Global.checkpoints` is wrong, the one specified by `Global.pretrain_weights` will be loaded.
W
WenmuZhou 已提交
89

W
WenmuZhou 已提交
90
### 3.3 Eval
W
WenmuZhou 已提交
91

W
WenmuZhou 已提交
92
The table uses [TEDS(Tree-Edit-Distance-based Similarity)](https://github.com/ibm-aur-nlp/PubTabNet/tree/master/src) as the evaluation metric of the model. Before the model evaluation, the three models in the pipeline need to be exported as inference models (we have provided them), and the gt for evaluation needs to be prepared. Examples of gt are as follows:
W
WenmuZhou 已提交
93
```json
W
WenmuZhou 已提交
94
{"PMC4289340_004_00.png": [
95 96
  ["<html>", "<body>", "<table>", "<thead>", "<tr>", "<td>", "</td>", "<td>", "</td>", "<td>", "</td>", "</tr>", "</thead>", "<tbody>", "<tr>", "<td>", "</td>", "<td>", "</td>", "<td>", "</td>", "</tr>",  "</tbody>", "</table>", "</body>", "</html>"],
  [[1, 4, 29, 13], [137, 4, 161, 13], [215, 4, 236, 13], [1, 17, 30, 27], [137, 17, 147, 27], [215, 17, 225, 27]],
W
WenmuZhou 已提交
97 98
  [["<b>", "F", "e", "a", "t", "u", "r", "e", "</b>"], ["<b>", "G", "b", "3", " ", "+", "</b>"], ["<b>", "G", "b", "3", " ", "-", "</b>"], ["<b>", "P", "a", "t", "i", "e", "n", "t", "s", "</b>"], ["6", "2"], ["4", "5"]]
]}
W
WenmuZhou 已提交
99 100 101 102 103 104 105 106
```
In gt json, the key is the image name, the value is the corresponding gt, and gt is a list composed of four items, and each item is
1. HTML string list of table structure
2. The coordinates of each cell (not including the empty text in the cell)
3. The text information in each cell (not including the empty text in the cell)

Use the following command to evaluate. After the evaluation is completed, the teds indicator will be output.
```python
W
WenmuZhou 已提交
107
cd PaddleOCR/ppstructure
108
python3 table/eval_table.py --det_model_dir=path/to/det_model_dir --rec_model_dir=path/to/rec_model_dir --table_model_dir=path/to/table_model_dir --image_dir=../doc/table/1.png --rec_char_dict_path=../ppocr/utils/dict/table_dict.txt --table_char_dict_path=../ppocr/utils/dict/table_structure_dict.txt --det_limit_side_len=736 --det_limit_type=min --gt_path=path/to/gt.json
W
WenmuZhou 已提交
109 110
```

W
WenmuZhou 已提交
111 112
If the PubLatNet eval dataset is used, it will be output
```bash
W
WenmuZhou 已提交
113
teds: 93.32
W
WenmuZhou 已提交
114
```
W
WenmuZhou 已提交
115

W
WenmuZhou 已提交
116
### 3.4 Inference
W
WenmuZhou 已提交
117 118

```python
W
WenmuZhou 已提交
119
cd PaddleOCR/ppstructure
文幕地方's avatar
文幕地方 已提交
120
python3 table/predict_table.py --det_model_dir=path/to/det_model_dir --rec_model_dir=path/to/rec_model_dir --table_model_dir=path/to/table_model_dir --image_dir=../doc/table/1.png --rec_char_dict_path=../ppocr/utils/dict/table_dict.txt --table_char_dict_path=../ppocr/utils/dict/table_structure_dict.txt --det_limit_side_len=736 --det_limit_type=min --output ../output/table
W
WenmuZhou 已提交
121
```
M
MissPenguin 已提交
122
After running, the excel sheet of each picture will be saved in the directory specified by the output field
W
WenmuZhou 已提交
123 124 125

Reference
1. https://github.com/ibm-aur-nlp/PubTabNet
126
2. https://arxiv.org/pdf/1911.10683