add PP-Structurev2 to hubserving

deploy/hubserving/readme.md

@@ -59,6 +59,7 @@ pip3 install paddlehub==2.1.0 --upgrade -i https://mirror.baidu.com/pypi/simple
 检测模型：./inference/ch_PP-OCRv3_det_infer/
 识别模型：./inference/ch_PP-OCRv3_rec_infer/
 方向分类器：./inference/ch_ppocr_mobile_v2.0_cls_infer/
+版面分析模型：
 表格结构识别模型：./inference/en_ppocr_mobile_v2.0_table_structure_infer/
 ```  
 
@@ -172,7 +173,7 @@ hub serving start -c deploy/hubserving/ocr_system/config.json
 ## 3. 发送预测请求
 配置好服务端，可使用以下命令发送预测请求，获取预测结果:  
 
-```python tools/test_hubserving.py server_url image_path```  
+```python tools/test_hubserving.py --server_url=server_url --image_dir=image_path```  
 
 需要给脚本传递2个参数：  
 - **server_url**：服务地址，格式为  

deploy/hubserving/readme_en.md

@@ -61,6 +61,7 @@ Before installing the service module, you need to prepare the inference model an
 text detection model: ./inference/ch_PP-OCRv3_det_infer/
 text recognition model: ./inference/ch_PP-OCRv3_rec_infer/
 text angle classifier: ./inference/ch_ppocr_mobile_v2.0_cls_infer/
+layout parse model:
 tanle recognition: ./inference/en_ppocr_mobile_v2.0_table_structure_infer/
 ```  
 
@@ -177,7 +178,7 @@ hub serving start -c deploy/hubserving/ocr_system/config.json
 ## 3. Send prediction requests
 After the service starts, you can use the following command to send a prediction request to obtain the prediction result:  
 ```shell
-python tools/test_hubserving.py server_url image_path
+python tools/test_hubserving.py --server_url=server_url --image_dir=image_path
 ```  
 
 Two parameters need to be passed to the script:

deploy/hubserving/structure_system/module.py

@@ -119,7 +119,7 @@ def predict(self, images=[], paths=[]):
                 all_results.append([])
                 continue
             starttime = time.time()
-            res = self.table_sys(img)
+            res, _ = self.table_sys(img)
             elapse = time.time() - starttime
             logger.info("Predict time: {}".format(elapse))
 
@@ -144,6 +144,6 @@ def serving_method(self, images, **kwargs):
 if __name__ == '__main__':
     structure_system = StructureSystem()
     structure_system._initialize()
-    image_path = ['./doc/table/1.png']
+    image_path = ['./ppstructure/docs/table/1.png']
     res = structure_system.predict(paths=image_path)
     print(res)

deploy/hubserving/structure_system/params.py

@@ -23,8 +23,10 @@ def read_params():
     cfg = table_read_params()
 
     # params for layout parser model
-    cfg.layout_path_model = 'lp://PubLayNet/ppyolov2_r50vd_dcn_365e_publaynet/config'
-    cfg.layout_label_map = None
+    cfg.layout_model_dir = ''
+    cfg.layout_dict_path = './ppocr/utils/dict/layout_publaynet_dict.txt'
+    cfg.layout_score_threshold = 0.5
+    cfg.layout_nms_threshold = 0.5
 
     cfg.mode = 'structure'
     cfg.output = './output'

deploy/hubserving/structure_table/module.py

@@ -118,11 +118,11 @@ def predict(self, images=[], paths=[]):
                 all_results.append([])
                 continue
             starttime = time.time()
-            pred_html = self.table_sys(img)
+            res, _ = self.table_sys(img)
             elapse = time.time() - starttime
             logger.info("Predict time: {}".format(elapse))
 
-            all_results.append({'html': pred_html})
+            all_results.append({'html': res['html']})
         return all_results
 
     @serving
@@ -138,6 +138,6 @@ def serving_method(self, images, **kwargs):
 if __name__ == '__main__':
     table_system = TableSystem()
     table_system._initialize()
-    image_path = ['./doc/table/table.jpg']
+    image_path = ['./ppstructure/docs/table/table.jpg']
     res = table_system.predict(paths=image_path)
     print(res)

doc/doc_en/quickstart_en.md

@@ -3,14 +3,14 @@
 **Note:** This tutorial mainly introduces the usage of PP-OCR series models, please refer to [PP-Structure Quick Start](../../ppstructure/docs/quickstart_en.md) for the quick use of document analysis related functions.
 
 - [1. Installation](#1-installation)
-    - [1.1 Install PaddlePaddle](#11-install-paddlepaddle)
-    - [1.2 Install PaddleOCR Whl Package](#12-install-paddleocr-whl-package)
+  - [1.1 Install PaddlePaddle](#11-install-paddlepaddle)
+  - [1.2 Install PaddleOCR Whl Package](#12-install-paddleocr-whl-package)
 - [2. Easy-to-Use](#2-easy-to-use)
-    - [2.1 Use by Command Line](#21-use-by-command-line)
-      - [2.1.1 Chinese and English Model](#211-chinese-and-english-model)
-      - [2.1.2 Multi-language Model](#212-multi-language-model)
-    - [2.2 Use by Code](#22-use-by-code)
-      - [2.2.1 Chinese & English Model and Multilingual Model](#221-chinese--english-model-and-multilingual-model)
+  - [2.1 Use by Command Line](#21-use-by-command-line)
+    - [2.1.1 Chinese and English Model](#211-chinese-and-english-model)
+    - [2.1.2 Multi-language Model](#212-multi-language-model)
+  - [2.2 Use by Code](#22-use-by-code)
+    - [2.2.1 Chinese & English Model and Multilingual Model](#221-chinese--english-model-and-multilingual-model)
 - [3. Summary](#3-summary)
 
 
@@ -51,12 +51,6 @@ pip install "paddleocr>=2.0.1" # Recommend to use version 2.0.1+
 
   Reference: [Solve shapely installation on windows](https://stackoverflow.com/questions/44398265/install-shapely-oserror-winerror-126-the-specified-module-could-not-be-found)
 
-- **For layout analysis users**, run the following command to install **Layout-Parser**
-
-  ```bash
-  pip3 install -U https://paddleocr.bj.bcebos.com/whl/layoutparser-0.0.0-py3-none-any.whl
-  ```
-
 <a name="2-easy-to-use"></a>
 
 ## 2. Easy-to-Use

ppstructure/docs/installation.md

@@ -1,8 +1,7 @@
 - [快速安装](#快速安装)
   - [1. PaddlePaddle 和 PaddleOCR](#1-paddlepaddle-和-paddleocr)
   - [2. 安装其他依赖](#2-安装其他依赖)
-    - [2.1 版面分析所需  Layout-Parser](#21-版面分析所需--layout-parser)
-    - [2.2  VQA所需依赖](#22--vqa所需依赖)
+    - [2.1  VQA所需依赖](#21--vqa所需依赖)
 
 # 快速安装
 
@@ -12,14 +11,7 @@
 
 ## 2. 安装其他依赖
 
-### 2.1 版面分析所需  Layout-Parser
-
-Layout-Parser 可通过如下命令安装
-
-```bash
-pip3 install -U https://paddleocr.bj.bcebos.com/whl/layoutparser-0.0.0-py3-none-any.whl
-```
-### 2.2  VQA所需依赖
+### 2.1  VQA所需依赖
 * paddleocr
 
 ```bash

ppstructure/docs/quickstart.md

@@ -1,21 +1,21 @@
 # PP-Structure 快速开始
 
-- [1. 安装依赖包](#1)
-- [2. 便捷使用](#2)
-    - [2.1 命令行使用](#21)
-        - [2.1.1 版面分析+表格识别](#211)
-        - [2.1.2 版面分析](#212)
-        - [2.1.3 表格识别](#213)
-        - [2.1.4 DocVQA](#214)
-    - [2.2 代码使用](#22)
-        - [2.2.1 版面分析+表格识别](#221)
-        - [2.2.2 版面分析](#222)
-        - [2.2.3 表格识别](#223)
-        - [2.2.4 DocVQA](#224)
-    - [2.3 返回结果说明](#23)
-        - [2.3.1 版面分析+表格识别](#231)
-        - [2.3.2 DocVQA](#232)
-    - [2.4 参数说明](#24)
+- [1. 安装依赖包](#1-安装依赖包)
+- [2. 便捷使用](#2-便捷使用)
+  - [2.1 命令行使用](#21-命令行使用)
+    - [2.1.1 版面分析+表格识别](#211-版面分析表格识别)
+    - [2.1.2 版面分析](#212-版面分析)
+    - [2.1.3 表格识别](#213-表格识别)
+    - [2.1.4 DocVQA](#214-docvqa)
+  - [2.2 代码使用](#22-代码使用)
+    - [2.2.1 版面分析+表格识别](#221-版面分析表格识别)
+    - [2.2.2 版面分析](#222-版面分析)
+    - [2.2.3 表格识别](#223-表格识别)
+    - [2.2.4 DocVQA](#224-docvqa)
+  - [2.3 返回结果说明](#23-返回结果说明)
+    - [2.3.1 版面分析+表格识别](#231-版面分析表格识别)
+    - [2.3.2 DocVQA](#232-docvqa)
+  - [2.4 参数说明](#24-参数说明)
 
 
 <a name="1"></a>
@@ -24,8 +24,6 @@
 ```bash
 # 安装 paddleocr，推荐使用2.5+版本
 pip3 install "paddleocr>=2.5"
-# 安装 版面分析依赖包layoutparser（如不需要版面分析功能，可跳过）
-pip3 install -U https://paddleocr.bj.bcebos.com/whl/layoutparser-0.0.0-py3-none-any.whl
 # 安装 DocVQA依赖包paddlenlp（如不需要DocVQA功能，可跳过）
 pip install paddlenlp
 

ppstructure/docs/quickstart_en.md

@@ -1,21 +1,21 @@
 # PP-Structure Quick Start
 
-- [1. Install package](#1)
-- [2. Use](#2)
-    - [2.1 Use by command line](#21)
-        - [2.1.1 layout analysis + table recognition](#211)
-        - [2.1.2 layout analysis](#212)
-        - [2.1.3 table recognition](#213)
-        - [2.1.4 DocVQA](#214)
-    - [2.2 Use by code](#22)
-        - [2.2.1 layout analysis + table recognition](#221)
-        - [2.2.2 layout analysis](#222)
-        - [2.2.3 table recognition](#223)
-        - [2.2.4 DocVQA](#224)
-    - [2.3 Result description](#23)
-        - [2.3.1 layout analysis + table recognition](#231)
-        - [2.3.2 DocVQA](#232)
-    - [2.4 Parameter Description](#24)
+- [1. Install package](#1-install-package)
+- [2. Use](#2-use)
+  - [2.1 Use by command line](#21-use-by-command-line)
+    - [2.1.1 layout analysis + table recognition](#211-layout-analysis--table-recognition)
+    - [2.1.2 layout analysis](#212-layout-analysis)
+    - [2.1.3 table recognition](#213-table-recognition)
+    - [2.1.4 DocVQA](#214-docvqa)
+  - [2.2 Use by code](#22-use-by-code)
+    - [2.2.1 layout analysis + table recognition](#221-layout-analysis--table-recognition)
+    - [2.2.2 layout analysis](#222-layout-analysis)
+    - [2.2.3 table recognition](#223-table-recognition)
+    - [2.2.4 DocVQA](#224-docvqa)
+  - [2.3 Result description](#23-result-description)
+    - [2.3.1 layout analysis + table recognition](#231-layout-analysis--table-recognition)
+    - [2.3.2 DocVQA](#232-docvqa)
+  - [2.4 Parameter Description](#24-parameter-description)
 
 
 <a name="1"></a>
@@ -24,8 +24,6 @@
 ```bash
 # Install paddleocr, version 2.5+ is recommended
 pip3 install "paddleocr>=2.5"
-# Install layoutparser (if you do not use the layout analysis, you can skip it)
-pip3 install -U https://paddleocr.bj.bcebos.com/whl/layoutparser-0.0.0-py3-none-any.whl
 # Install the DocVQA dependency package paddlenlp (if you do not use the DocVQA, you can skip it)
 pip install paddlenlp
 

ppstructure/predict_system.py

@@ -43,6 +43,7 @@
 class StructureSystem(object):
     def __init__(self, args):
         self.mode = args.mode
+        self.recovery = args.recovery
         if self.mode == 'structure':
             if not args.show_log:
                 logger.setLevel(logging.INFO)
@@ -110,7 +111,7 @@ def __call__(self, img, return_ocr_result_in_table=False):
                         time_dict['rec'] += table_time_dict['rec']
                 else:
                     if self.text_system is not None:
-                        if args.recovery:
+                        if self.recovery:
                             wht_im = np.ones(ori_im.shape, dtype=ori_im.dtype)
                             wht_im[y1:y2, x1:x2, :] = roi_img
                             filter_boxes, filter_rec_res, ocr_time_dict = self.text_system(
@@ -133,7 +134,7 @@ def __call__(self, img, return_ocr_result_in_table=False):
                             for token in style_token:
                                 if token in rec_str:
                                     rec_str = rec_str.replace(token, '')
-                            if not args.recovery:
+                            if not self.recovery:
                                 box += [x1, y1]
                             res.append({
                                 'text': rec_str,

ppstructure/table/predict_table.py

@@ -101,7 +101,7 @@ def __call__(self, img, return_ocr_result_in_table=False):
         start = time.time()
 
         structure_res, elapse = self._structure(copy.deepcopy(img))
-        result['cell_bbox'] = structure_res[1]
+        result['cell_bbox'] = structure_res[1].tolist()
         time_dict['table'] = elapse
 
         dt_boxes, rec_res, det_elapse, rec_elapse = self._ocr(

ppstructure/utility.py

@@ -38,14 +38,17 @@ def init_args():
     parser.add_argument(
         "--layout_dict_path",
         type=str,
-        default="../ppocr/utils/dict/layout_pubalynet_dict.txt")
+        default="../ppocr/utils/dict/layout_publaynet_dict.txt")
     parser.add_argument(
         "--layout_score_threshold",
         type=float,
         default=0.5,
         help="Threshold of score.")
     parser.add_argument(
-        "--layout_nms_threshold", type=float, default=0.5, help="Threshold of nms.")
+        "--layout_nms_threshold",
+        type=float,
+        default=0.5,
+        help="Threshold of nms.")
     # params for vqa
     parser.add_argument("--vqa_algorithm", type=str, default='LayoutXLM')
     parser.add_argument("--ser_model_dir", type=str)