Merge branch 'dygraph' of https://github.com/PaddlePaddle/PaddleOCR into dygraph_rc

PPOCRLabel/PPOCRLabel.py

@@ -64,6 +64,7 @@ from libs.colorDialog import ColorDialog
 from libs.toolBar import ToolBar
 from libs.ustr import ustr
 from libs.hashableQListWidgetItem import HashableQListWidgetItem
+from libs.editinlist import EditInList
 
 __appname__ = 'PPOCRLabel'
 
@@ -147,7 +148,7 @@ class MainWindow(QMainWindow, WindowMixin):
         self.prevLabelText = getStr('tempLabel')
         self.model = 'paddle'
         self.PPreader = None
-        self.autoSaveNum = 10
+        self.autoSaveNum = 5
 
         ################# file list  ###############
         self.fileListWidget = QListWidget()
@@ -201,12 +202,12 @@ class MainWindow(QMainWindow, WindowMixin):
 
         ################## label list ####################
         # Create and add a widget for showing current label items
-        self.labelList = QListWidget()
+        self.labelList = EditInList()
         labelListContainer = QWidget()
         labelListContainer.setLayout(listLayout)
         self.labelList.itemActivated.connect(self.labelSelectionChanged)
         self.labelList.itemSelectionChanged.connect(self.labelSelectionChanged)
-        self.labelList.itemDoubleClicked.connect(self.editLabel)
+        self.labelList.clicked.connect(self.labelList.item_clicked)
         # Connect to itemChanged to detect checkbox changes.
         self.labelList.itemChanged.connect(self.labelItemChanged)
         self.labelListDock = QDockWidget(getStr('recognitionResult'),self)
@@ -316,7 +317,7 @@ class MainWindow(QMainWindow, WindowMixin):
         self.scrollArea = scroll
         self.canvas.scrollRequest.connect(self.scrollRequest)
 
-        self.canvas.newShape.connect(self.newShape)
+        self.canvas.newShape.connect(partial(self.newShape, False))
         self.canvas.shapeMoved.connect(self.updateBoxlist)  # self.setDirty
         self.canvas.selectionChanged.connect(self.shapeSelectionChanged)
         self.canvas.drawingPolygon.connect(self.toggleDrawingSensitive)
@@ -354,13 +355,9 @@ class MainWindow(QMainWindow, WindowMixin):
         quit = action(getStr('quit'), self.close,
                       'Ctrl+Q', 'quit', getStr('quitApp'))
 
-        open = action(getStr('openFile'), self.openFile,
-                      'Ctrl+O', 'open', getStr('openFileDetail'))
-
         opendir = action(getStr('openDir'), self.openDirDialog,
                          'Ctrl+u', 'open', getStr('openDir'))
 
-
         save = action(getStr('save'), self.saveFile,
                       'Ctrl+V', 'verify', getStr('saveDetail'), enabled=False)
 
@@ -506,7 +503,7 @@ class MainWindow(QMainWindow, WindowMixin):
         self.drawSquaresOption.triggered.connect(self.toogleDrawSquare)
 
         # Store actions for further handling.
-        self.actions = struct(save=save,  open=open,  resetAll=resetAll, deleteImg=deleteImg,
+        self.actions = struct(save=save,  resetAll=resetAll, deleteImg=deleteImg,
                               lineColor=color1, create=create, delete=delete, edit=edit, copy=copy,
                               saveRec=saveRec, singleRere=singleRere,AutoRec=AutoRec,reRec=reRec,
                               createMode=createMode, editMode=editMode,
@@ -515,7 +512,7 @@ class MainWindow(QMainWindow, WindowMixin):
                               fitWindow=fitWindow, fitWidth=fitWidth,
                               zoomActions=zoomActions, saveLabel=saveLabel,
                               fileMenuActions=(
-                                  open, opendir, saveLabel,  resetAll, quit),
+                                  opendir, saveLabel,  resetAll, quit),
                               beginner=(), advanced=(),
                               editMenu=(createpoly, edit, copy, delete,singleRere,
                                         None, color1, self.drawSquaresOption),
@@ -537,11 +534,6 @@ class MainWindow(QMainWindow, WindowMixin):
             labelList=labelMenu)
 
 
-        # Sync single class mode from PR#106
-        self.singleClassMode = QAction(getStr('singleClsMode'), self)
-        self.singleClassMode.setShortcut("Ctrl+Shift+S")
-        self.singleClassMode.setCheckable(True)
-        self.singleClassMode.setChecked(settings.get(SETTING_SINGLE_CLASS, False))
         self.lastLabel = None
         # Add option to enable/disable labels being displayed at the top of bounding boxes
         self.displayLabelOption = QAction(getStr('displayLabel'), self)
@@ -550,12 +542,18 @@ class MainWindow(QMainWindow, WindowMixin):
         self.displayLabelOption.setChecked(settings.get(SETTING_PAINT_LABEL, False))
         self.displayLabelOption.triggered.connect(self.togglePaintLabelsOption)
 
+        self.labelDialogOption = QAction(getStr('labelDialogOption'), self)
+        self.labelDialogOption.setShortcut("Ctrl+Shift+L")
+        self.labelDialogOption.setCheckable(True)
+        self.labelDialogOption.setChecked(settings.get(SETTING_PAINT_LABEL, False))
+        self.labelDialogOption.triggered.connect(self.speedChoose)
+
         addActions(self.menus.file,
                    (opendir, None, saveLabel, saveRec, None, resetAll, deleteImg, quit))
 
         addActions(self.menus.help, (showSteps, showInfo))
         addActions(self.menus.view, (
-            self.displayLabelOption, # labels,
+            self.displayLabelOption, self.labelDialogOption,
              None,
             hideAll, showAll, None,
             zoomIn, zoomOut, zoomOrg, None,
@@ -1062,6 +1060,7 @@ class MainWindow(QMainWindow, WindowMixin):
 
     def labelSelectionChanged(self):
         item = self.currentItem()
+        self.labelList.scrollToItem(item, QAbstractItemView.EnsureVisible)
         if item and self.canvas.editing():
             self._noSelectionSlot = True
             self.canvas.selectShape(self.itemsToShapes[item])
@@ -1069,6 +1068,7 @@ class MainWindow(QMainWindow, WindowMixin):
 
     def boxSelectionChanged(self):
         item = self.currentBox()
+        self.BoxList.scrollToItem(item, QAbstractItemView.EnsureVisible)
         if item and self.canvas.editing():
             self._noSelectionSlot = True
             self.canvas.selectShape(self.itemsToShapesbox[item])
@@ -1089,7 +1089,7 @@ class MainWindow(QMainWindow, WindowMixin):
             # self.actions.save.setEnabled(True)
 
     # Callback functions:
-    def newShape(self):
+    def newShape(self, value=True):
         """Pop-up and give focus to the label editor.
 
         position MUST be in global coordinates.
@@ -1098,12 +1098,11 @@ class MainWindow(QMainWindow, WindowMixin):
             self.labelDialog = LabelDialog(
                 parent=self, listItem=self.labelHist)
 
-        # Sync single class mode from PR#106
+        if value:
-        if self.singleClassMode.isChecked() and self.lastLabel:
-            text = self.lastLabel
-        else:
             text = self.labelDialog.popUp(text=self.prevLabelText)
             self.lastLabel = text
+        else:
+            text = self.prevLabelText
 
         if text is not None:
             self.prevLabelText = self.stringBundle.getString('tempLabel')
@@ -1364,7 +1363,6 @@ class MainWindow(QMainWindow, WindowMixin):
             else:
                 settings[SETTING_LAST_OPEN_DIR] = ''
 
-            settings[SETTING_SINGLE_CLASS] = self.singleClassMode.isChecked()
             settings[SETTING_PAINT_LABEL] = self.displayLabelOption.isChecked()
             settings[SETTING_DRAW_SQUARE] = self.drawSquaresOption.isChecked()
             settings.save()
@@ -1497,35 +1495,6 @@ class MainWindow(QMainWindow, WindowMixin):
             print('file name in openNext is ',filename)
             self.loadFile(filename)
         
-    def openFile(self, _value=False):
-        if not self.mayContinue():
-            return
-        path = os.path.dirname(ustr(self.filePath)) if self.filePath else '.'
-        formats = ['*.%s' % fmt.data().decode("ascii").lower() for fmt in QImageReader.supportedImageFormats()]
-        filters = "Image & Label files (%s)" % ' '.join(formats + ['*%s' % LabelFile.suffix])
-        filename = QFileDialog.getOpenFileName(self, '%s - Choose Image or Label file' % __appname__, path, filters)
-        if filename:
-            if isinstance(filename, (tuple, list)):
-                filename = filename[0]
-            self.loadFile(filename)
-            # print('filename in openfile is ', self.filePath)
-        self.filePath = None
-        self.fileListWidget.clear()
-        self.iconlist.clear()
-        self.mImgList = [filename]
-        self.openNextImg()
-        if self.validFilestate(filename) is True:
-            item = QListWidgetItem(newIcon('done'), filename)
-            self.setClean()
-        elif self.validFilestate(filename) is None:
-            item = QListWidgetItem(newIcon('close'), filename)
-        else:
-            item = QListWidgetItem(newIcon('close'), filename)
-            self.setDirty()
-        self.fileListWidget.addItem(filename)
-        self.additems5(None)
-        print('opened image is', filename)
-        
     def updateFileListIcon(self, filename):
         pass
 
@@ -1963,6 +1932,16 @@ class MainWindow(QMainWindow, WindowMixin):
 
         QMessageBox.information(self, "Information", "Cropped images has been saved in "+str(crop_img_dir))
 
+    def speedChoose(self):
+        if self.labelDialogOption.isChecked():
+            self.canvas.newShape.disconnect()
+            self.canvas.newShape.connect(partial(self.newShape, True))
+
+        else:
+            self.canvas.newShape.disconnect()
+            self.canvas.newShape.connect(partial(self.newShape, False))
+
+
 def inverted(color):
     return QColor(*[255 - v for v in color.getRgb()])
 

PPOCRLabel/README.md

@@ -2,19 +2,27 @@ English | [简体中文](README_ch.md)
 
 # PPOCRLabel
 
-PPOCRLabel is a semi-automatic graphic annotation tool suitable for OCR field. It is written in python3 and pyqt5, supporting rectangular box annotation and four-point annotation modes. Annotations can be directly used for the training of PPOCR detection and recognition models.
+PPOCRLabel is a semi-automatic graphic annotation tool suitable for OCR field, with built-in PPOCR model to automatically detect and re-recognize data. It is written in python3 and pyqt5, supporting rectangular box annotation and four-point annotation modes. Annotations can be directly used for the training of PPOCR detection and recognition models.
 
 <img src="./data/gif/steps_en.gif" width="100%"/>
 
 ### Recent Update
 
+- 2021.1.11: Optimize the labeling experience (by [edencfc](https://github.com/edencfc)),
+  - Users can choose whether to pop up the label input dialog after drawing the detection box in "View - Pop-up Label Input Dialog". 
+  - The recognition result scrolls synchronously when users click related detection box.
+  - Click to modify the recognition result.(If you can't change the result, please switch to the system default input method, or switch back to the original input method again)
 - 2020.12.18: Support re-recognition of a single label box (by [ninetailskim](https://github.com/ninetailskim) ), perfect shortcut keys.
 
+### TODO:
+- Lock box mode: For the same scene data, the size and position of the locked detection box can be transferred between different pictures.
+- Experience optimization: Add undo, batch operation include move, copy, delete and so on, optimize the annotation process.
+
 ## Installation
 
 ### 1. Install PaddleOCR
 
-Refer to [PaddleOCR installation document](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/doc/doc_ch/installation.md) to prepare PaddleOCR
+PaddleOCR models has been built in PPOCRLabel, please refer to [PaddleOCR installation document](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/doc/doc_ch/installation.md) to prepare PaddleOCR and make sure it works.
 
 ### 2. Install PPOCRLabel
 
@@ -60,7 +68,7 @@ python3 PPOCRLabel.py
 
    4.1 Click 'Create RectBox' or press 'W' in English keyboard mode to draw a new rectangle detection box. Click and release left mouse to select a region to annotate the text area.
 
-   4.2 Press 'P' to enter four-point labeling mode which enables you to create any four-point shape by clicking four points with the left mouse button in succession and DOUBLE CLICK the left mouse as the signal of labeling completion.
+   4.2 Press 'Q' to enter four-point labeling mode which enables you to create any four-point shape by clicking four points with the left mouse button in succession and DOUBLE CLICK the left mouse as the signal of labeling completion.
 
 5. After the marking frame is drawn, the user clicks "OK", and the detection frame will be pre-assigned a "TEMPORARY" label.
 
@@ -72,7 +80,7 @@ python3 PPOCRLabel.py
 
 9. Click "Delete Image" and the image will be deleted to the recycle bin.
 
-10. Labeling result: the user can save manually through the menu "File - Save Label", while the program will also save automatically after every 10 images confirmed by the user.the manually checked label will be stored in *Label.txt* under the opened picture folder.
+10. Labeling result: the user can save manually through the menu "File - Save Label", while the program will also save automatically after every 5 images confirmed by the user.the manually checked label will be stored in *Label.txt* under the opened picture folder.
     Click "PaddleOCR"-"Save Recognition Results" in the menu bar, the recognition training data of such pictures will be saved in the *crop_img* folder, and the recognition label will be saved in *rec_gt.txt*<sup>[4]</sup>.
 
 ### Note
@@ -88,7 +96,7 @@ Therefore, if the recognition result has been manually changed before, it may ch
 
 |   File name   |                         Description                          |
 | :-----------: | :----------------------------------------------------------: |
-|   Label.txt   | The detection label file can be directly used for PPOCR detection model training. After the user saves 10 label results, the file will be automatically saved. It will also be written when the user closes the application or changes the file folder. |
+|   Label.txt   | The detection label file can be directly used for PPOCR detection model training. After the user saves 5 label results, the file will be automatically saved. It will also be written when the user closes the application or changes the file folder. |
 | fileState.txt | The picture status file save the image in the current folder that has been manually confirmed by the user. |
 |  Cache.cach   |    Cache files to save the results of model recognition.     |
 |  rec_gt.txt   | The recognition label file, which can be directly used for PPOCR identification model training, is generated after the user clicks on the menu bar "File"-"Save recognition result". |
@@ -124,6 +132,15 @@ Therefore, if the recognition result has been manually changed before, it may ch
 
 - Custom model: The model trained by users can be replaced by modifying PPOCRLabel.py in [PaddleOCR class instantiation](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/PPOCRLabel/PPOCRLabel.py#L110) referring [Custom Model Code](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/doc/doc_en/whl_en.md#use-custom-model)
 
+### Save
+
+PPOCRLabel supports three ways to save Label.txt
+
+- Automatically save: When it detects that the user has manually checked 5 pictures, the program automatically writes the annotations into Label.txt. The user can change the value of ``self.autoSaveNum`` in ``PPOCRLabel.py`` to set the number of images to be automatically saved after confirmation.
+- Manual save: Click "File-Save Marking Results" to manually save the label.
+- Close application save
+
+
 ### Export partial recognition results
 
 For some data that are difficult to recognize, the recognition results will not be exported by **unchecking** the corresponding tags in the recognition results checkbox.

PPOCRLabel/README_ch.md

@@ -2,18 +2,30 @@
 
 # PPOCRLabel
 
-PPOCRLabel是一款适用于OCR领域的半自动化图形标注工具，使用python3和pyqt5编写，支持矩形框标注和四点标注模式，导出格式可直接用于PPOCR检测和识别模型的训练。
+PPOCRLabel是一款适用于OCR领域的半自动化图形标注工具，内置PPOCR模型对数据自动标注和重新识别。使用python3和pyqt5编写，支持矩形框标注和四点标注模式，导出格式可直接用于PPOCR检测和识别模型的训练。
 
 <img src="./data/gif/steps.gif" width="100%"/>
 
 #### 近期更新
 
-- 2020.12.18: 支持对单个标记框进行重新识别（by [ninetailskim](https://github.com/ninetailskim) ），完善快捷键。
+- 2021.1.11：优化标注体验（by [edencfc](https://github.com/edencfc)）：
+  - 用户可在“视图 - 弹出标记输入框”选择在画完检测框后标记输入框是否弹出。
+  - 识别结果与检测框同步滚动。
+  - 识别结果更改为单击修改。（如果无法修改，请切换为系统自带输入法，或再次切回原输入法）
+- 2020.12.18： 支持对单个标记框进行重新识别（by [ninetailskim](https://github.com/ninetailskim)），完善快捷键。
+
+#### 尽请期待
+
+- 锁定框模式：针对同一场景数据，被锁定的检测框的大小与位置能在不同图片之间传递。
+- 体验优化：增加撤销操作，批量移动、复制、删除等功能。优化标注流程。
+
+如果您对以上内容感兴趣或对完善工具有不一样的想法，欢迎加入我们的队伍与我们共同开发
+
 
 ## 安装
 
 ### 1. 安装PaddleOCR
-参考[PaddleOCR安装文档](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/doc/doc_ch/installation.md)准备好PaddleOCR
+PPOCRLabel内置PaddleOCR模型，故请参考[PaddleOCR安装文档](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/doc/doc_ch/installation.md)准备好PaddleOCR，并确保PaddleOCR安装成功。
 
 ### 2. 安装PPOCRLabel
 #### Windows + Anaconda
@@ -49,13 +61,13 @@ python3 PPOCRLabel.py --lang ch
 1. 安装与运行：使用上述命令安装与运行程序。
 2. 打开文件夹：在菜单栏点击 “文件” - "打开目录" 选择待标记图片的文件夹<sup>[1]</sup>.
 3. 自动标注：点击 ”自动标注“，使用PPOCR超轻量模型对图片文件名前图片状态<sup>[2]</sup>为 “X” 的图片进行自动标注。
-4. 手动标注：点击 “矩形标注”（推荐直接在英文模式下点击键盘中的 “W”)，用户可对当前图片中模型未检出的部分进行手动绘制标记框。点击键盘P，则使用四点标注模式（或点击“编辑” - “四点标注”），用户依次点击4个点后，双击左键表示标注完成。
+4. 手动标注：点击 “矩形标注”（推荐直接在英文模式下点击键盘中的 “W”)，用户可对当前图片中模型未检出的部分进行手动绘制标记框。点击键盘Q，则使用四点标注模式（或点击“编辑” - “四点标注”），用户依次点击4个点后，双击左键表示标注完成。
 5. 标记框绘制完成后，用户点击 “确认”，检测框会先被预分配一个 “待识别” 标签。
 6. 重新识别：将图片中的所有检测画绘制/调整完成后，点击 “重新识别”，PPOCR模型会对当前图片中的**所有检测框**重新识别<sup>[3]</sup>。
 7. 内容更改：双击识别结果，对不准确的识别结果进行手动更改。
 8. 确认标记：点击 “确认”，图片状态切换为 “√”，跳转至下一张（此时不会直接将结果写入文件）。
 9. 删除：点击 “删除图像”，图片将会被删除至回收站。
-10. 保存结果：用户可以通过菜单中“文件-保存标记结果”手动保存，同时程序也会在用户每确认10张图片后自动保存一次。手动确认过的标记将会被存放在所打开图片文件夹下的*Label.txt*中。在菜单栏点击 “文件” - "保存识别结果"后，会将此类图片的识别训练数据保存在*crop_img*文件夹下，识别标签保存在*rec_gt.txt*中<sup>[4]</sup>。
+10. 保存结果：用户可以通过菜单中“文件-保存标记结果”手动保存，同时程序也会在用户每确认5张图片后自动保存一次。手动确认过的标记将会被存放在所打开图片文件夹下的*Label.txt*中。在菜单栏点击 “文件” - "保存识别结果"后，会将此类图片的识别训练数据保存在*crop_img*文件夹下，识别标签保存在*rec_gt.txt*中<sup>[4]</sup>。
 
 ### 注意
 
@@ -69,7 +81,7 @@ python3 PPOCRLabel.py --lang ch
 
 |    文件名     |                             说明                             |
 | :-----------: | :----------------------------------------------------------: |
-|   Label.txt   | 检测标签，可直接用于PPOCR检测模型训练。用户每保存10张检测结果后，程序会进行自动写入。当用户关闭应用程序或切换文件路径后同样会进行写入。 |
+|   Label.txt   | 检测标签，可直接用于PPOCR检测模型训练。用户每保存5张检测结果后，程序会进行自动写入。当用户关闭应用程序或切换文件路径后同样会进行写入。 |
 | fileState.txt | 图片状态标记文件，保存当前文件夹下已经被用户手动确认过的图片名称。 |
 |  Cache.cach   |              缓存文件，保存模型自动识别的结果。              |
 |  rec_gt.txt   | 识别标签。可直接用于PPOCR识别模型训练。需用户手动点击菜单栏“文件” - "保存识别结果"后产生。 |
@@ -104,6 +116,14 @@ python3 PPOCRLabel.py --lang ch
 
  - 自定义模型：用户可根据[自定义模型代码使用](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/doc/doc_ch/whl.md#%E8%87%AA%E5%AE%9A%E4%B9%89%E6%A8%A1%E5%9E%8B)，通过修改PPOCRLabel.py中针对[PaddleOCR类的实例化](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/PPOCRLabel/PPOCRLabel.py#L110)替换成自己训练的模型。
 
+### 保存方式
+
+PPOCRLabel支持三种保存方式：
+
+- 程序自动保存：当检测到用户手动确认过5张图片后，程序自动将标记结果写入Label.txt中。其中用户可通过更改```PPOCRLabel.py```中的```self.autoSaveNum```的数值设置确认几张图片后进行自动保存。
+- 手动保存：点击“文件 - 保存标记结果”手动保存标记。
+- 关闭应用程序保存
+
 ### 导出部分识别结果
 
 针对部分难以识别的数据，通过在识别结果的复选框中**取消勾选**相应的标记，其识别结果不会被导出。
@@ -115,7 +135,7 @@ python3 PPOCRLabel.py --lang ch
   
 - PPOCRLabel**不支持对中文文件名**的图片进行自动标注。
 
-- 针对Linux用户：：如果您在打开软件过程中出现**objc[XXXXX]**开头的错误，证明您的opencv版本太高，建议安装4.2版本：
+- 针对Linux用户：如果您在打开软件过程中出现**objc[XXXXX]**开头的错误，证明您的opencv版本太高，建议安装4.2版本：
     ```
     pip install opencv-python==4.2.0.32
     ```
@@ -129,6 +149,7 @@ python3 PPOCRLabel.py --lang ch
     ```
     pip install opencv-contrib-python-headless
     ```
+
 ### 参考资料
 
 1.[Tzutalin. LabelImg. Git code (2015)](https://github.com/tzutalin/labelImg)

PPOCRLabel/libs/editinlist.py

@@ -0,0 +1,31 @@
+import sys, time
+from PyQt5 import QtWidgets
+from PyQt5.QtGui import *
+from PyQt5.QtCore import *
+from PyQt5.QtWidgets import *
+
+class EditInList(QListWidget):
+    def __init__(self):
+        super(EditInList,self).__init__()
+        # click to edit
+        self.clicked.connect(self.item_clicked)  
+
+    def item_clicked(self, modelindex: QModelIndex) -> None:
+        self.edited_item = self.currentItem()
+        self.closePersistentEditor(self.edited_item)
+        item = self.item(modelindex.row())
+        # time.sleep(0.2)
+        self.edited_item = item
+        self.openPersistentEditor(item)
+        # time.sleep(0.2)
+        self.editItem(item)
+
+    def mouseDoubleClickEvent(self, event):
+        # close edit
+        for i in range(self.count()):
+            self.closePersistentEditor(self.item(i))
+
+    def leaveEvent(self, event):
+        # close edit
+        for i in range(self.count()):
+            self.closePersistentEditor(self.item(i))

PPOCRLabel/requirements/requirements-linux-python3.txt

@@ -1,2 +0,0 @@
-pyqt5==5.10.1
-lxml==4.2.4

PPOCRLabel/resources/strings/strings-zh-CN.properties

@@ -42,7 +42,7 @@ zoomin=放大画面
 info=信息
 openAnnotation=开启标签
 prevImgDetail=上一个图像
-fitWidth=缩放到跟当前画面一样宽
+fitWidth=缩放到当前画面宽度
 zoomout=缩小画面
 changeSavedAnnotationDir=更改保存标签文件的预设目录
 nextImgDetail=下一个图像
@@ -96,3 +96,4 @@ hideBox=隐藏所有标注
 showBox=显示所有标注
 saveLabel=保存标记结果
 singleRe=重识别此区块
+labelDialogOption=弹出标记输入框

PPOCRLabel/resources/strings/strings.properties

@@ -96,3 +96,4 @@ hideBox=Hide All Box
 showBox=Show All Box
 saveLabel=Save Label
 singleRe=Re-recognition RectBox
+labelDialogOption=Pop-up Label Input Dialog

README.md

@@ -173,7 +173,7 @@ This project is released under <a href="https://github.com/PaddlePaddle/PaddleOC
 We welcome all the contributions to PaddleOCR and appreciate for your feedback very much.
 
 - Many thanks to [Khanh Tran](https://github.com/xxxpsyduck) and [Karl Horky](https://github.com/karlhorky) for contributing and revising the English documentation.
-- Many thanks to [zhangxin](https://github.com/ZhangXinNan) for contributing the new visualize function、add .gitgnore and discard set PYTHONPATH manually.
+- Many thanks to [zhangxin](https://github.com/ZhangXinNan) for contributing the new visualize function、add .gitignore and discard set PYTHONPATH manually.
 - Many thanks to [lyl120117](https://github.com/lyl120117) for contributing the code for printing the network structure.
 - Thanks [xiangyubo](https://github.com/xiangyubo) for contributing the handwritten Chinese OCR datasets.
 - Thanks [authorfu](https://github.com/authorfu) for contributing Android demo  and [xiadeye](https://github.com/xiadeye) contributing iOS demo, respectively.

README_ch.md

@@ -8,7 +8,7 @@ PaddleOCR同时支持动态图与静态图两种编程范式
 - 静态图版本：develop分支
 
 **近期更新**
-- 2020.12.28 [FAQ](./doc/doc_ch/FAQ.md)新增5个高频问题，总数137个，每周一都会更新，欢迎大家持续关注。
+- 2021.1.18 [FAQ](./doc/doc_ch/FAQ.md)新增5个高频问题，总数152个，每周一都会更新，欢迎大家持续关注。
 - 2020.12.15 更新数据合成工具[Style-Text](./StyleText/README_ch.md)，可以批量合成大量与目标场景类似的图像，在多个场景验证，效果明显提升。
 - 2020.11.25 更新半自动标注工具[PPOCRLabel](./PPOCRLabel/README_ch.md)，辅助开发者高效完成标注任务，输出格式与PP-OCR训练任务完美衔接。
 - 2020.9.22 更新PP-OCR技术文章，https://arxiv.org/abs/2009.09941
@@ -101,8 +101,8 @@ PaddleOCR同时支持动态图与静态图两种编程范式
 - [效果展示](#效果展示)
 - FAQ
     - [【精选】OCR精选10个问题](./doc/doc_ch/FAQ.md)
-    - [【理论篇】OCR通用31个问题](./doc/doc_ch/FAQ.md)
+    - [【理论篇】OCR通用32个问题](./doc/doc_ch/FAQ.md)
-    - [【实战篇】PaddleOCR实战96个问题](./doc/doc_ch/FAQ.md)
+    - [【实战篇】PaddleOCR实战110个问题](./doc/doc_ch/FAQ.md)
 - [技术交流群](#欢迎加入PaddleOCR技术交流群)
 - [参考文献](./doc/doc_ch/reference.md)
 - [许可证书](#许可证书)
@@ -149,7 +149,7 @@ PP-OCR是一个实用的超轻量OCR系统。主要由DB文本检测[2]、检测
 
 
 - 非常感谢 [Khanh Tran](https://github.com/xxxpsyduck) 和 [Karl Horky](https://github.com/karlhorky) 贡献修改英文文档
-- 非常感谢 [zhangxin](https://github.com/ZhangXinNan)([Blog](https://blog.csdn.net/sdlypyzq)) 贡献新的可视化方式、添加.gitgnore、处理手动设置PYTHONPATH环境变量的问题
+- 非常感谢 [zhangxin](https://github.com/ZhangXinNan)([Blog](https://blog.csdn.net/sdlypyzq)) 贡献新的可视化方式、添加.gitignore、处理手动设置PYTHONPATH环境变量的问题
 - 非常感谢 [lyl120117](https://github.com/lyl120117) 贡献打印网络结构的代码
 - 非常感谢 [xiangyubo](https://github.com/xiangyubo) 贡献手写中文OCR数据集
 - 非常感谢 [authorfu](https://github.com/authorfu) 贡献Android和[xiadeye](https://github.com/xiadeye) 贡献IOS的demo代码

StyleText/README.md

@@ -52,7 +52,7 @@ If you save the model in another location, please modify the address of the mode
 
 ```
 bg_generator:
-  pretrain: style_text_rec/bg_generator
+  pretrain: style_text_models/bg_generator
 ...
 text_generator:
   pretrain: style_text_models/text_generator

configs/det/det_mv3_db.yml

@@ -102,6 +102,7 @@ Train:
     drop_last: False
     batch_size_per_card: 16
     num_workers: 8
+    use_shared_memory: False
 
 Eval:
   dataset:
@@ -129,3 +130,4 @@ Eval:
     drop_last: False
     batch_size_per_card: 1 # must be 1
     num_workers: 8
+    use_shared_memory: False

configs/rec/rec_icdar15_train.yml

@@ -76,6 +76,7 @@ Train:
     batch_size_per_card: 256
     drop_last: True
     num_workers: 8
+    use_shared_memory: False
 
 Eval:
   dataset:
@@ -96,3 +97,4 @@ Eval:
     drop_last: False
     batch_size_per_card: 256
     num_workers: 4
+    use_shared_memory: False

deploy/cpp_infer/CMakeLists.txt

@@ -138,12 +138,22 @@ endif()
 
 # Note: libpaddle_inference_api.so/a must put before libpaddle_fluid.so/a
 if(WITH_STATIC_LIB)
-  set(DEPS
+  if(WIN32)
-      ${PADDLE_LIB}/paddle/lib/libpaddle_fluid${CMAKE_STATIC_LIBRARY_SUFFIX})
+    set(DEPS
+        ${PADDLE_LIB}/paddle/lib/paddle_fluid${CMAKE_STATIC_LIBRARY_SUFFIX})
+  else()
+    set(DEPS
+        ${PADDLE_LIB}/paddle/lib/libpaddle_fluid${CMAKE_STATIC_LIBRARY_SUFFIX})
+  endif()
 else()
-  set(DEPS
+  if(WIN32)
-      ${PADDLE_LIB}/paddle/lib/libpaddle_fluid${CMAKE_SHARED_LIBRARY_SUFFIX})
+    set(DEPS
-endif()
+        ${PADDLE_LIB}/paddle/lib/paddle_fluid${CMAKE_SHARED_LIBRARY_SUFFIX})
+  else()
+    set(DEPS
+        ${PADDLE_LIB}/paddle/lib/libpaddle_fluid${CMAKE_SHARED_LIBRARY_SUFFIX})
+  endif()
+endif(WITH_STATIC_LIB)
 
 if (NOT WIN32)
     set(DEPS ${DEPS}

deploy/cpp_infer/include/config.h

@@ -62,6 +62,10 @@ public:
     this->cls_thresh = stod(config_map_["cls_thresh"]);
 
     this->visualize = bool(stoi(config_map_["visualize"]));
+
+    this->use_tensorrt = bool(stoi(config_map_["use_tensorrt"]));
+
+    this->use_fp16 = bool(stod(config_map_["use_fp16"]));
   }
 
   bool use_gpu = false;
@@ -96,6 +100,10 @@ public:
 
   bool visualize = true;
 
+  bool use_tensorrt = false;
+
+  bool use_fp16 = false;
+
   void PrintConfigInfo();
 
 private:

deploy/cpp_infer/include/ocr_cls.h

@@ -39,7 +39,8 @@ public:
   explicit Classifier(const std::string &model_dir, const bool &use_gpu,
                       const int &gpu_id, const int &gpu_mem,
                       const int &cpu_math_library_num_threads,
-                      const bool &use_mkldnn, const double &cls_thresh) {
+                      const bool &use_mkldnn, const double &cls_thresh,
+                      const bool &use_tensorrt, const bool &use_fp16) {
     this->use_gpu_ = use_gpu;
     this->gpu_id_ = gpu_id;
     this->gpu_mem_ = gpu_mem;
@@ -47,6 +48,8 @@ public:
     this->use_mkldnn_ = use_mkldnn;
 
     this->cls_thresh = cls_thresh;
+    this->use_tensorrt_ = use_tensorrt;
+    this->use_fp16_ = use_fp16;
 
     LoadModel(model_dir);
   }
@@ -69,7 +72,8 @@ private:
   std::vector<float> mean_ = {0.5f, 0.5f, 0.5f};
   std::vector<float> scale_ = {1 / 0.5f, 1 / 0.5f, 1 / 0.5f};
   bool is_scale_ = true;
-
+  bool use_tensorrt_ = false;
+  bool use_fp16_ = false;
   // pre-process
   ClsResizeImg resize_op_;
   Normalize normalize_op_;

deploy/cpp_infer/include/ocr_det.h

@@ -44,8 +44,8 @@ public:
                       const bool &use_mkldnn, const int &max_side_len,
                       const double &det_db_thresh,
                       const double &det_db_box_thresh,
-                      const double &det_db_unclip_ratio,
+                      const double &det_db_unclip_ratio, const bool &visualize,
-                      const bool &visualize) {
+                      const bool &use_tensorrt, const bool &use_fp16) {
     this->use_gpu_ = use_gpu;
     this->gpu_id_ = gpu_id;
     this->gpu_mem_ = gpu_mem;
@@ -59,6 +59,8 @@ public:
     this->det_db_unclip_ratio_ = det_db_unclip_ratio;
 
     this->visualize_ = visualize;
+    this->use_tensorrt_ = use_tensorrt;
+    this->use_fp16_ = use_fp16;
 
     LoadModel(model_dir);
   }
@@ -85,6 +87,8 @@ private:
   double det_db_unclip_ratio_ = 2.0;
 
   bool visualize_ = true;
+  bool use_tensorrt_ = false;
+  bool use_fp16_ = false;
 
   std::vector<float> mean_ = {0.485f, 0.456f, 0.406f};
   std::vector<float> scale_ = {1 / 0.229f, 1 / 0.224f, 1 / 0.225f};

deploy/cpp_infer/include/ocr_rec.h

@@ -41,12 +41,15 @@ public:
   explicit CRNNRecognizer(const std::string &model_dir, const bool &use_gpu,
                           const int &gpu_id, const int &gpu_mem,
                           const int &cpu_math_library_num_threads,
-                          const bool &use_mkldnn, const string &label_path) {
+                          const bool &use_mkldnn, const string &label_path,
+                          const bool &use_tensorrt, const bool &use_fp16) {
     this->use_gpu_ = use_gpu;
     this->gpu_id_ = gpu_id;
     this->gpu_mem_ = gpu_mem;
     this->cpu_math_library_num_threads_ = cpu_math_library_num_threads;
     this->use_mkldnn_ = use_mkldnn;
+    this->use_tensorrt_ = use_tensorrt;
+    this->use_fp16_ = use_fp16;
 
     this->label_list_ = Utility::ReadDict(label_path);
     this->label_list_.insert(this->label_list_.begin(),
@@ -76,7 +79,8 @@ private:
   std::vector<float> mean_ = {0.5f, 0.5f, 0.5f};
   std::vector<float> scale_ = {1 / 0.5f, 1 / 0.5f, 1 / 0.5f};
   bool is_scale_ = true;
-
+  bool use_tensorrt_ = false;
+  bool use_fp16_ = false;
   // pre-process
   CrnnResizeImg resize_op_;
   Normalize normalize_op_;

deploy/cpp_infer/include/preprocess_op.h

@@ -47,18 +47,20 @@ public:
 class ResizeImgType0 {
 public:
   virtual void Run(const cv::Mat &img, cv::Mat &resize_img, int max_size_len,
-                   float &ratio_h, float &ratio_w);
+                   float &ratio_h, float &ratio_w, bool use_tensorrt);
 };
 
 class CrnnResizeImg {
 public:
   virtual void Run(const cv::Mat &img, cv::Mat &resize_img, float wh_ratio,
+                   bool use_tensorrt = false,
                    const std::vector<int> &rec_image_shape = {3, 32, 320});
 };
 
 class ClsResizeImg {
 public:
   virtual void Run(const cv::Mat &img, cv::Mat &resize_img,
+                   bool use_tensorrt = false,
                    const std::vector<int> &rec_image_shape = {3, 48, 192});
 };
 

deploy/cpp_infer/src/main.cpp

@@ -54,18 +54,20 @@ int main(int argc, char **argv) {
                  config.gpu_mem, config.cpu_math_library_num_threads,
                  config.use_mkldnn, config.max_side_len, config.det_db_thresh,
                  config.det_db_box_thresh, config.det_db_unclip_ratio,
-                 config.visualize);
+                 config.visualize, config.use_tensorrt, config.use_fp16);
 
   Classifier *cls = nullptr;
   if (config.use_angle_cls == true) {
     cls = new Classifier(config.cls_model_dir, config.use_gpu, config.gpu_id,
                          config.gpu_mem, config.cpu_math_library_num_threads,
-                         config.use_mkldnn, config.cls_thresh);
+                         config.use_mkldnn, config.cls_thresh,
+                         config.use_tensorrt, config.use_fp16);
   }
 
   CRNNRecognizer rec(config.rec_model_dir, config.use_gpu, config.gpu_id,
                      config.gpu_mem, config.cpu_math_library_num_threads,
-                     config.use_mkldnn, config.char_list_file);
+                     config.use_mkldnn, config.char_list_file,
+                     config.use_tensorrt, config.use_fp16);
 
   auto start = std::chrono::system_clock::now();
   std::vector<std::vector<std::vector<int>>> boxes;
@@ -75,11 +77,11 @@ int main(int argc, char **argv) {
   auto end = std::chrono::system_clock::now();
   auto duration =
       std::chrono::duration_cast<std::chrono::microseconds>(end - start);
-  std::cout << "花费了"
+  std::cout << "Cost  "
             << double(duration.count()) *
                    std::chrono::microseconds::period::num /
                    std::chrono::microseconds::period::den
-            << "秒" << std::endl;
+            << "s" << std::endl;
 
   return 0;
 }

deploy/cpp_infer/src/ocr_cls.cpp

@@ -25,7 +25,7 @@ cv::Mat Classifier::Run(cv::Mat &img) {
   int index = 0;
   float wh_ratio = float(img.cols) / float(img.rows);
 
-  this->resize_op_.Run(img, resize_img, cls_image_shape);
+  this->resize_op_.Run(img, resize_img, this->use_tensorrt_, cls_image_shape);
 
   this->normalize_op_.Run(&resize_img, this->mean_, this->scale_,
                           this->is_scale_);
@@ -76,6 +76,13 @@ void Classifier::LoadModel(const std::string &model_dir) {
 
   if (this->use_gpu_) {
     config.EnableUseGpu(this->gpu_mem_, this->gpu_id_);
+    if (this->use_tensorrt_) {
+      config.EnableTensorRtEngine(
+          1 << 20, 10, 3,
+          this->use_fp16_ ? paddle_infer::Config::Precision::kHalf
+                          : paddle_infer::Config::Precision::kFloat32,
+          false, false);
+    }
   } else {
     config.DisableGpu();
     if (this->use_mkldnn_) {

deploy/cpp_infer/src/ocr_det.cpp

@@ -24,10 +24,13 @@ void DBDetector::LoadModel(const std::string &model_dir) {
 
   if (this->use_gpu_) {
     config.EnableUseGpu(this->gpu_mem_, this->gpu_id_);
-    //     config.EnableTensorRtEngine(
+    if (this->use_tensorrt_) {
-    //           1 << 20, 1, 3,
+      config.EnableTensorRtEngine(
-    //           AnalysisConfig::Precision::kFloat32,
+          1 << 20, 10, 3,
-    //           false, false);
+          this->use_fp16_ ? paddle_infer::Config::Precision::kHalf
+                          : paddle_infer::Config::Precision::kFloat32,
+          false, false);
+    }
   } else {
     config.DisableGpu();
     if (this->use_mkldnn_) {
@@ -58,7 +61,8 @@ void DBDetector::Run(cv::Mat &img,
   cv::Mat srcimg;
   cv::Mat resize_img;
   img.copyTo(srcimg);
-  this->resize_op_.Run(img, resize_img, this->max_side_len_, ratio_h, ratio_w);
+  this->resize_op_.Run(img, resize_img, this->max_side_len_, ratio_h, ratio_w,
+                       this->use_tensorrt_);
 
   this->normalize_op_.Run(&resize_img, this->mean_, this->scale_,
                           this->is_scale_);

deploy/cpp_infer/src/ocr_rec.cpp

@@ -33,7 +33,7 @@ void CRNNRecognizer::Run(std::vector<std::vector<std::vector<int>>> boxes,
 
     float wh_ratio = float(crop_img.cols) / float(crop_img.rows);
 
-    this->resize_op_.Run(crop_img, resize_img, wh_ratio);
+    this->resize_op_.Run(crop_img, resize_img, wh_ratio, this->use_tensorrt_);
 
     this->normalize_op_.Run(&resize_img, this->mean_, this->scale_,
                             this->is_scale_);
@@ -76,7 +76,7 @@ void CRNNRecognizer::Run(std::vector<std::vector<std::vector<int>>> boxes,
           float(*std::max_element(&predict_batch[n * predict_shape[2]],
                                   &predict_batch[(n + 1) * predict_shape[2]]));
 
-      if (argmax_idx > 0 && (not(i > 0 && argmax_idx == last_index))) {
+      if (argmax_idx > 0 && (!(i > 0 && argmax_idx == last_index))) {
         score += max_value;
         count += 1;
         str_res.push_back(label_list_[argmax_idx]);
@@ -99,6 +99,13 @@ void CRNNRecognizer::LoadModel(const std::string &model_dir) {
 
   if (this->use_gpu_) {
     config.EnableUseGpu(this->gpu_mem_, this->gpu_id_);
+    if (this->use_tensorrt_) {
+      config.EnableTensorRtEngine(
+          1 << 20, 10, 3,
+          this->use_fp16_ ? paddle_infer::Config::Precision::kHalf
+                          : paddle_infer::Config::Precision::kFloat32,
+          false, false);
+    }
   } else {
     config.DisableGpu();
     if (this->use_mkldnn_) {

deploy/cpp_infer/src/preprocess_op.cpp

@@ -60,7 +60,8 @@ void Normalize::Run(cv::Mat *im, const std::vector<float> &mean,
 }
 
 void ResizeImgType0::Run(const cv::Mat &img, cv::Mat &resize_img,
-                         int max_size_len, float &ratio_h, float &ratio_w) {
+                         int max_size_len, float &ratio_h, float &ratio_w,
+                         bool use_tensorrt) {
   int w = img.cols;
   int h = img.rows;
 
@@ -89,14 +90,19 @@ void ResizeImgType0::Run(const cv::Mat &img, cv::Mat &resize_img,
     resize_w = 32;
   else
     resize_w = (resize_w / 32) * 32;
-
+  if (!use_tensorrt) {
-  cv::resize(img, resize_img, cv::Size(resize_w, resize_h));
+    cv::resize(img, resize_img, cv::Size(resize_w, resize_h));
-
+    ratio_h = float(resize_h) / float(h);
-  ratio_h = float(resize_h) / float(h);
+    ratio_w = float(resize_w) / float(w);
-  ratio_w = float(resize_w) / float(w);
+  } else {
+    cv::resize(img, resize_img, cv::Size(640, 640));
+    ratio_h = float(640) / float(h);
+    ratio_w = float(640) / float(w);
+  }
 }
 
 void CrnnResizeImg::Run(const cv::Mat &img, cv::Mat &resize_img, float wh_ratio,
+                        bool use_tensorrt,
                         const std::vector<int> &rec_image_shape) {
   int imgC, imgH, imgW;
   imgC = rec_image_shape[0];
@@ -111,12 +117,27 @@ void CrnnResizeImg::Run(const cv::Mat &img, cv::Mat &resize_img, float wh_ratio,
     resize_w = imgW;
   else
     resize_w = int(ceilf(imgH * ratio));
-
+  if (!use_tensorrt) {
-  cv::resize(img, resize_img, cv::Size(resize_w, imgH), 0.f, 0.f,
+    cv::resize(img, resize_img, cv::Size(resize_w, imgH), 0.f, 0.f,
-             cv::INTER_LINEAR);
+               cv::INTER_LINEAR);
+    cv::copyMakeBorder(resize_img, resize_img, 0, 0, 0,
+                       int(imgW - resize_img.cols), cv::BORDER_CONSTANT,
+                       {127, 127, 127});
+  } else {
+    int k = int(img.cols * 32 / img.rows);
+    if (k >= 100) {
+      cv::resize(img, resize_img, cv::Size(100, 32), 0.f, 0.f,
+                 cv::INTER_LINEAR);
+    } else {
+      cv::resize(img, resize_img, cv::Size(k, 32), 0.f, 0.f, cv::INTER_LINEAR);
+      cv::copyMakeBorder(resize_img, resize_img, 0, 0, 0, int(100 - k),
+                         cv::BORDER_CONSTANT, {127, 127, 127});
+    }
+  }
 }
 
 void ClsResizeImg::Run(const cv::Mat &img, cv::Mat &resize_img,
+                       bool use_tensorrt,
                        const std::vector<int> &rec_image_shape) {
   int imgC, imgH, imgW;
   imgC = rec_image_shape[0];
@@ -130,11 +151,15 @@ void ClsResizeImg::Run(const cv::Mat &img, cv::Mat &resize_img,
   else
     resize_w = int(ceilf(imgH * ratio));
 
-  cv::resize(img, resize_img, cv::Size(resize_w, imgH), 0.f, 0.f,
+  if (!use_tensorrt) {
-             cv::INTER_LINEAR);
+    cv::resize(img, resize_img, cv::Size(resize_w, imgH), 0.f, 0.f,
-  if (resize_w < imgW) {
+               cv::INTER_LINEAR);
-    cv::copyMakeBorder(resize_img, resize_img, 0, 0, 0, imgW - resize_w,
+    if (resize_w < imgW) {
-                       cv::BORDER_CONSTANT, cv::Scalar(0, 0, 0));
+      cv::copyMakeBorder(resize_img, resize_img, 0, 0, 0, imgW - resize_w,
+                         cv::BORDER_CONSTANT, cv::Scalar(0, 0, 0));
+    }
+  } else {
+    cv::resize(img, resize_img, cv::Size(100, 32), 0.f, 0.f, cv::INTER_LINEAR);
   }
 }
 

deploy/cpp_infer/tools/config.txt

@@ -24,3 +24,7 @@ char_list_file ../../ppocr/utils/ppocr_keys_v1.txt
 # show the detection results
 visualize 1
 
+# use_tensorrt
+use_tensorrt 0
+use_fp16   0
+