Merge branch 'PaddlePaddle:dygraph' into dygraph

PPOCRLabel/README.md

@@ -8,6 +8,8 @@ PPOCRLabel is a semi-automatic graphic annotation tool suitable for OCR field, w
 
 ### Recent Update
 
+- 2022.01：（by [PeterH0323](https://github.com/peterh0323) ）
+  - Improve user experience: prompt for the number of files and labels, optimize interaction, and fix bugs such as only use CPU when inference
 - 2021.11.17:
   - Support install and start PPOCRLabel through the whl package (by [d2623587501](https://github.com/d2623587501))
   - Dataset segmentation: Divide the annotation file into training, verification and testing parts (refer to section 3.5 below, by [MrCuiHao](https://github.com/MrCuiHao))
@@ -110,7 +112,7 @@ python PPOCRLabel.py
 
 6. Click 're-Recognition', model will rewrite ALL recognition results in ALL detection box<sup>[3]</sup>.
 
-7. Double click the result in 'recognition result' list to manually change inaccurate recognition results.
+7. Single click the result in 'recognition result' list to manually change inaccurate recognition results.
 
 8. **Click "Check", the image status will switch to "√",then the program automatically jump to the next.**
 
@@ -143,15 +145,17 @@ python PPOCRLabel.py
 ### 3.1 Shortcut keys
 
 | Shortcut keys            | Description                                      |
-| ------------------------ | ------------------------------------------------ |
+|--------------------------|--------------------------------------------------|
 | Ctrl + Shift + R         | Re-recognize all the labels of the current image |
 | W                        | Create a rect box                                |
 | Q                        | Create a four-points box                         |
+| X                        | Rotate the box anti-clockwise                    |
+| C                        | Rotate the box clockwise                         |
 | Ctrl + E                 | Edit label of the selected box                   |
 | Ctrl + R                 | Re-recognize the selected box                    |
 | Ctrl + C                 | Copy and paste the selected box                  |
 | Ctrl + Left Mouse Button | Multi select the label box                       |
-| Backspace                | Delete the selected box                          |
+| Alt + X                  | Delete the selected box                          |
 | Ctrl + V                 | Check image                                      |
 | Ctrl + Shift + d         | Delete image                                     |
 | D                        | Next image                                       |

PPOCRLabel/README_ch.md

@@ -8,6 +8,8 @@ PPOCRLabel是一款适用于OCR领域的半自动化图形标注工具，内置P
 
 #### 近期更新
 
+- 2022.01：（by [PeterH0323](https://github.com/peterh0323) ）
+  - 提升用户体验：新增文件与标记数目提示、优化交互、修复gpu使用等问题
 - 2021.11.17：
   - 新增支持通过whl包安装和启动PPOCRLabel（by [d2623587501](https://github.com/d2623587501)）
   - 标注数据集切分：对标注数据进行训练、验证与测试集划分（参考下方3.5节，by [MrCuiHao](https://github.com/MrCuiHao)）
@@ -102,7 +104,7 @@ python PPOCRLabel.py --lang ch
 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。
+7. 内容更改：单击识别结果，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。**
 9. 删除：点击 “删除图像”，图片将会被删除至回收站。
 10. 导出结果：用户可以通过菜单中“文件-导出标记结果”手动导出，同时也可以点击“文件 - 自动导出标记结果”开启自动导出。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” - "导出识别结果"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>。
@@ -131,23 +133,25 @@ python PPOCRLabel.py --lang ch
 
 ### 3.1 快捷键
 
-| 快捷键           | 说明                         |
-| ---------------- | ---------------------------- |
-| Ctrl + shift + R | 对当前图片的所有标记重新识别 |
-| W                | 新建矩形框                   |
-| Q                | 新建四点框                   |
-| Ctrl + E         | 编辑所选框标签               |
-| Ctrl + R         | 重新识别所选标记             |
+| 快捷键              | 说明             |
+|------------------|----------------|
+| Ctrl + shift + R | 对当前图片的所有标记重新识别  |
+| W                | 新建矩形框                  |
+| Q                | 新建四点框                  |
+| X                | 框逆时针旋转                |
+| C                | 框顺时针旋转                |
+| Ctrl + E         | 编辑所选框标签              |
+| Ctrl + R         | 重新识别所选标记            |
 | Ctrl + C         | 复制并粘贴选中的标记框       |
-| Ctrl + 鼠标左键  | 多选标记框                   |
-| Backspace        | 删除所选框                   |
-| Ctrl + V         | 确认本张图片标记             |
-| Ctrl + Shift + d | 删除本张图片                 |
-| D                | 下一张图片                   |
-| A                | 上一张图片                   |
-| Ctrl++           | 缩小                         |
-| Ctrl--           | 放大                         |
-| ↑→↓←             | 移动标记框                   |
+| Ctrl + 鼠标左键    | 多选标记框                 |
+| Alt + X          | 删除所选框                 |
+| Ctrl + V         | 确认本张图片标记           |
+| Ctrl + Shift + d | 删除本张图片               |
+| D                | 下一张图片                |
+| A                | 上一张图片                |
+| Ctrl++           | 缩小                     |
+| Ctrl--           | 放大                     |
+| ↑→↓←             | 移动标记框                |
 
 ### 3.2 内置模型
 

PPOCRLabel/combobox.py

-# Copyright (c) <2015-Present> Tzutalin
-# Copyright (C) 2013  MIT, Computer Science and Artificial Intelligence Laboratory. Bryan Russell, Antonio Torralba,
-# William T. Freeman. Permission is hereby granted, free of charge, to any person obtaining a copy of this software and
-# associated documentation files (the "Software"), to deal in the Software without restriction, including without
-# limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the
-# Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
-# The above copyright notice and this permission notice shall be included in all copies or substantial portions of
-# the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT
-# NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT
-# SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF
-# CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
-# THE SOFTWARE.
-
-import sys
-try:
-    from PyQt5.QtWidgets import QWidget, QHBoxLayout, QComboBox
-except ImportError:
-    # needed for py3+qt4
-    # Ref:
-    # http://pyqt.sourceforge.net/Docs/PyQt4/incompatible_apis.html
-    # http://stackoverflow.com/questions/21217399/pyqt4-qtcore-qvariant-object-instead-of-a-string
-    if sys.version_info.major >= 3:
-        import sip
-        sip.setapi('QVariant', 2)
-    from PyQt4.QtGui import QWidget, QHBoxLayout, QComboBox
-
-
-class ComboBox(QWidget):
-    def __init__(self, parent=None, items=[]):
-        super(ComboBox, self).__init__(parent)
-
-        layout = QHBoxLayout()
-        self.cb = QComboBox()
-        self.items = items
-        self.cb.addItems(self.items)
-
-        self.cb.currentIndexChanged.connect(parent.comboSelectionChanged)
-
-        layout.addWidget(self.cb)
-        self.setLayout(layout)
-
-    def update_items(self, items):
-        self.items = items
-
-        self.cb.clear()
-        self.cb.addItems(self.items)

PPOCRLabel/libs/autoDialog.py

@@ -6,6 +6,8 @@ except ImportError:
     from PyQt4.QtGui import *
     from PyQt4.QtCore import *
 
+import time
+import datetime
 import json
 import cv2
 import numpy as np
@@ -80,8 +82,9 @@ class AutoDialog(QDialog):
         self.parent = parent
         self.ocr = ocr
         self.mImgList = mImgList
+        self.lender = lenbar
         self.pb = QProgressBar()
-        self.pb.setRange(0, lenbar)
+        self.pb.setRange(0, self.lender)
         self.pb.setValue(0)
 
         layout = QVBoxLayout()
@@ -108,10 +111,16 @@ class AutoDialog(QDialog):
         self.thread_1.progressBarValue.connect(self.handleProgressBarSingal)
         self.thread_1.listValue.connect(self.handleListWidgetSingal)
         self.thread_1.endsignal.connect(self.handleEndsignalSignal)
+        self.time_start = time.time()  # save start time
 
     def handleProgressBarSingal(self, i):
         self.pb.setValue(i)
 
+        # calculate time left of auto labeling
+        avg_time = (time.time() - self.time_start) / i  # Use average time to prevent time fluctuations
+        time_left = str(datetime.timedelta(seconds=avg_time * (self.lender - i))).split(".")[0]  # Remove microseconds
+        self.setWindowTitle("PPOCRLabel  --  " + f"Time Left: {time_left}")  # show
+
     def handleListWidgetSingal(self, i):
         self.listWidget.addItem(i)
         titem = self.listWidget.item(self.listWidget.count() - 1)

PPOCRLabel/libs/canvas.py

@@ -11,19 +11,13 @@
 # CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
 # THE SOFTWARE.
 
-try:
-    from PyQt5.QtGui import *
-    from PyQt5.QtCore import *
-    from PyQt5.QtWidgets import *
-except ImportError:
-    from PyQt4.QtGui import *
-    from PyQt4.QtCore import *
-
-#from PyQt4.QtOpenGL import *
+import copy
 
+from PyQt5.QtCore import Qt, pyqtSignal, QPointF, QPoint
+from PyQt5.QtGui import QPainter, QBrush, QColor, QPixmap
+from PyQt5.QtWidgets import QWidget, QMenu, QApplication
 from libs.shape import Shape
 from libs.utils import distance
-import copy
 
 CURSOR_DEFAULT = Qt.ArrowCursor
 CURSOR_POINT = Qt.PointingHandCursor
@@ -31,8 +25,6 @@ CURSOR_DRAW = Qt.CrossCursor
 CURSOR_MOVE = Qt.ClosedHandCursor
 CURSOR_GRAB = Qt.OpenHandCursor
 
-# class Canvas(QGLWidget):
-
 
 class Canvas(QWidget):
     zoomRequest = pyqtSignal(int)
@@ -129,7 +121,6 @@ class Canvas(QWidget):
     def selectedVertex(self):
         return self.hVertex is not None
 
-
     def mouseMoveEvent(self, ev):
         """Update line with last point and current coordinates."""
         pos = self.transformPos(ev.pos())
@@ -333,7 +324,6 @@ class Canvas(QWidget):
 
              self.movingShape = False
 
-
     def endMove(self, copy=False):
         assert self.selectedShapes and self.selectedShapesCopy
         assert len(self.selectedShapesCopy) == len(self.selectedShapes)
@@ -410,7 +400,6 @@ class Canvas(QWidget):
         self.selectionChanged.emit(shapes)
         self.update()
 
-
     def selectShapePoint(self, point, multiple_selection_mode):
         """Select the first shape created which contains this point."""
         if self.selectedVertex():  # A vertex is marked for selection.
@@ -494,7 +483,6 @@ class Canvas(QWidget):
         else:
             shape.moveVertexBy(index, shiftPos)
 
-
     def boundedMoveShape(self, shapes, pos):
         if type(shapes).__name__ != 'list': shapes = [shapes]
         if self.outOfPixmap(pos):
@@ -515,6 +503,7 @@ class Canvas(QWidget):
         if dp:
             for shape in shapes:
                 shape.moveBy(dp)
+                shape.close()
             self.prevPoint = pos
             return True
         return False
@@ -728,6 +717,31 @@ class Canvas(QWidget):
              self.moveOnePixel('Up')
         elif key == Qt.Key_Down and self.selectedShapes:
              self.moveOnePixel('Down')
+        elif key == Qt.Key_X and self.selectedShapes:
+            for i in range(len(self.selectedShapes)):
+                self.selectedShape = self.selectedShapes[i]
+                if self.rotateOutOfBound(0.01):
+                    continue
+                self.selectedShape.rotate(0.01)
+            self.shapeMoved.emit()
+            self.update()
+
+        elif key == Qt.Key_C and self.selectedShapes:
+            for i in range(len(self.selectedShapes)):
+                self.selectedShape = self.selectedShapes[i]
+                if self.rotateOutOfBound(-0.01):
+                    continue
+                self.selectedShape.rotate(-0.01)
+            self.shapeMoved.emit()
+            self.update()
+
+    def rotateOutOfBound(self, angle):
+        for shape in range(len(self.selectedShapes)):
+            self.selectedShape = self.selectedShapes[shape]
+            for i, p in enumerate(self.selectedShape.points):
+                if self.outOfPixmap(self.selectedShape.rotatePoint(p, angle)):
+                    return True
+            return False
 
     def moveOnePixel(self, direction):
         # print(self.selectedShape.points)

PPOCRLabel/libs/editinlist.py

-import sys, time
-from PyQt5 import QtWidgets
-from PyQt5.QtGui import *
-from PyQt5.QtCore import *
-from PyQt5.QtWidgets import *
+# !/usr/bin/env python
+# -*- coding: utf-8 -*-
+from PyQt5.QtCore import QModelIndex
+from PyQt5.QtWidgets import QListWidget
+
 
 class EditInList(QListWidget):
     def __init__(self):
-        super(EditInList,self).__init__()
-        # click to edit
-        self.clicked.connect(self.item_clicked)  
+        super(EditInList, self).__init__()
+        self.edited_item = None
+
+    def item_clicked(self, modelindex: QModelIndex):
+        try:
+            if self.edited_item is not None:
+                self.closePersistentEditor(self.edited_item)
+        except:
+            self.edited_item = self.currentItem()
 
-    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)
+        self.edited_item = self.item(modelindex.row())
+        self.openPersistentEditor(self.edited_item)
+        self.editItem(self.edited_item)
 
     def mouseDoubleClickEvent(self, event):
-        # close edit
-        for i in range(self.count()):
-            self.closePersistentEditor(self.item(i))
+        pass
 
     def leaveEvent(self, event):
         # close edit
         for i in range(self.count()):
-            self.closePersistentEditor(self.item(i))
\ No newline at end of file
+            self.closePersistentEditor(self.item(i))

PPOCRLabel/libs/shape.py

@@ -10,19 +10,14 @@
 # SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF
 # CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
 # THE SOFTWARE.
-#!/usr/bin/python
+# !/usr/bin/python
 # -*- coding: utf-8 -*-
+import math
+import sys
 
-
-try:
-    from PyQt5.QtGui import *
-    from PyQt5.QtCore import *
-except ImportError:
-    from PyQt4.QtGui import *
-    from PyQt4.QtCore import *
-
+from PyQt5.QtCore import QPointF
+from PyQt5.QtGui import QColor, QPen, QPainterPath, QFont
 from libs.utils import distance
-import sys
 
 DEFAULT_LINE_COLOR = QColor(0, 255, 0, 128)
 DEFAULT_FILL_COLOR = QColor(255, 0, 0, 128)
@@ -59,6 +54,8 @@ class Shape(object):
         self.difficult = difficult
         self.paintLabel = paintLabel
         self.locked = False
+        self.direction = 0
+        self.center = None
         self._highlightIndex = None
         self._highlightMode = self.NEAR_VERTEX
         self._highlightSettings = {
@@ -74,7 +71,24 @@ class Shape(object):
             # is used for drawing the pending line a different color.
             self.line_color = line_color
 
+    def rotate(self, theta):
+        for i, p in enumerate(self.points):
+            self.points[i] = self.rotatePoint(p, theta)
+        self.direction -= theta
+        self.direction = self.direction % (2 * math.pi)
+
+    def rotatePoint(self, p, theta):
+        order = p - self.center
+        cosTheta = math.cos(theta)
+        sinTheta = math.sin(theta)
+        pResx = cosTheta * order.x() + sinTheta * order.y()
+        pResy = - sinTheta * order.x() + cosTheta * order.y()
+        pRes = QPointF(self.center.x() + pResx, self.center.y() + pResy)
+        return pRes
+
     def close(self):
+        self.center = QPointF((self.points[0].x() + self.points[2].x()) / 2,
+                              (self.points[0].y() + self.points[2].y()) / 2)
         self._closed = True
 
     def reachMaxPoints(self):
@@ -83,7 +97,9 @@ class Shape(object):
         return False
 
     def addPoint(self, point):
-        if not self.reachMaxPoints():  # 4个点时发出close信号
+        if self.reachMaxPoints():
+            self.close()
+        else:
             self.points.append(point)
 
     def popPoint(self):
@@ -112,7 +128,7 @@ class Shape(object):
             # Uncommenting the following line will draw 2 paths
             # for the 1st vertex, and make it non-filled, which
             # may be desirable.
-            #self.drawVertex(vrtx_path, 0)
+            # self.drawVertex(vrtx_path, 0)
 
             for i, p in enumerate(self.points):
                 line_path.lineTo(p)
@@ -136,9 +152,9 @@ class Shape(object):
                     font.setPointSize(8)
                     font.setBold(True)
                     painter.setFont(font)
-                    if(self.label == None):
+                    if self.label is None:
                         self.label = ""
-                    if(min_y < MIN_Y_LABEL):
+                    if min_y < MIN_Y_LABEL:
                         min_y += MIN_Y_LABEL
                     painter.drawText(min_x, min_y, self.label)
 
@@ -198,6 +214,8 @@ class Shape(object):
     def copy(self):
         shape = Shape("%s" % self.label)
         shape.points = [p for p in self.points]
+        shape.center = self.center
+        shape.direction = self.direction
         shape.fill = self.fill
         shape.selected = self.selected
         shape._closed = self._closed

doc/doc_ch/models_list.md

@@ -50,7 +50,7 @@ PaddleOCR提供的可下载模型包括`推理模型`、`训练模型`、`预训
 |模型名称|模型简介|配置文件|推理模型大小|下载地址|
 | --- | --- | --- | --- | --- |
 |ch_PP-OCRv2_rec_slim|【最新】slim量化版超轻量模型，支持中英文、数字识别|[ch_PP-OCRv2_rec.yml](../../configs/rec/ch_PP-OCRv2/ch_PP-OCRv2_rec.yml)| 9M |[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_slim_quant_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_slim_quant_train.tar) |
-|ch_PP-OCRv2_rec|【最新】原始超轻量模型，支持中英文、数字识别|[ch_PP-OCRv2_rec.yml](../../configs/rec/ch_PP-OCRv2/ch_PP-OCRv2_rec.yml)|8.5M|[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_train.tar) |
+|ch_PP-OCRv2_rec|【最新】原始超轻量模型，支持中英文、数字识别|[ch_PP-OCRv2_rec_distillation.yml](../../configs/rec/ch_PP-OCRv2/ch_PP-OCRv2_rec_distillation.yml)|8.5M|[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_train.tar) |
 |ch_ppocr_mobile_slim_v2.0_rec|slim裁剪量化版超轻量模型，支持中英文、数字识别|[rec_chinese_lite_train_v2.0.yml](../../configs/rec/ch_ppocr_v2.0/rec_chinese_lite_train_v2.0.yml)| 6M |[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_slim_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_slim_train.tar) |
 |ch_ppocr_mobile_v2.0_rec|原始超轻量模型，支持中英文、数字识别|[rec_chinese_lite_train_v2.0.yml](../../configs/rec/ch_ppocr_v2.0/rec_chinese_lite_train_v2.0.yml)|5.2M|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_train.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_pre.tar) |
 |ch_ppocr_server_v2.0_rec|通用模型，支持中英文、数字识别|[rec_chinese_common_train_v2.0.yml](../../configs/rec/ch_ppocr_v2.0/rec_chinese_common_train_v2.0.yml)|94.8M|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_train.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_pre.tar) |

doc/doc_ch/thirdparty.md

@@ -16,22 +16,24 @@ PaddleOCR希望可以通过AI的力量助力任何一位有梦想的开发者实
 
 ### 1.1 基于PaddleOCR的社区项目
 
-- 【最新】 [FastOCRLabel](https://gitee.com/BaoJianQiang/FastOCRLabel)：完整的C#版本标注工具 (@ [包建强](https://gitee.com/BaoJianQiang) )
-
-#### 1.1.1 通用工具
-
-- [DangoOCR离线版](https://github.com/PantsuDango/DangoOCR)：通用型桌面级即时翻译工具 (@ [PantsuDango](https://github.com/PantsuDango))
-- [scr2txt](https://github.com/lstwzd/scr2txt)：截屏转文字工具 (@ [lstwzd](https://github.com/lstwzd))
-- [AI Studio项目](https://aistudio.baidu.com/aistudio/projectdetail/1054614?channelType=0&channel=0)：英文视频自动生成字幕( @ [叶月水狐](https://aistudio.baidu.com/aistudio/personalcenter/thirdview/322052))
-
-#### 1.1.2 垂类场景工具
-
-- [id_card_ocr](https://github.com/baseli/id_card_ocr)：身份证复印件识别(@ [baseli](https://github.com/baseli))
-- [Paddle_Table_Image_Reader](https://github.com/thunder95/Paddle_Table_Image_Reader)：能看懂表格图片的数据助手(@ [thunder95](https://github.com/thunder95]))
-
-#### 1.1.3 前后处理
-
-- [paddleOCRCorrectOutputs](https://github.com/yuranusduke/paddleOCRCorrectOutputs)：获取OCR识别结果的key-value(@ [yuranusduke](https://github.com/yuranusduke))
+| 类别     | 项目                                                         | 描述                       | 开发者                                                       |
+| -------- | ------------------------------------------------------------ | -------------------------- | ------------------------------------------------------------ |
+| 通用工具 | [FastOCRLabel](https://gitee.com/BaoJianQiang/FastOCRLabel)  | 完整的C#版本标注GUI        | [包建强](https://gitee.com/BaoJianQiang)                     |
+| 通用工具 | [DangoOCR离线版](https://github.com/PantsuDango/DangoOCR)    | 通用型桌面级即时翻译GUI    | [PantsuDango](https://github.com/PantsuDango)                |
+| 通用工具 | [scr2txt](https://github.com/lstwzd/scr2txt)                 | 截屏转文字GUI              | [lstwzd](https://github.com/lstwzd)                          |
+| 通用工具 | [ocr_sdk](https://github.com/mymagicpower/AIAS/blob/main/1_image_sdks/text_recognition/ocr_sdk)                 | OCR java SDK工具箱              | [Calvin](https://github.com/mymagicpower)                          |
+| 通用工具 | [iocr](https://github.com/mymagicpower/AIAS/blob/main/8_suite_hub/iocr)                 | IOCR 自定义模板识别(支持表格识别)              | [Calvin](https://github.com/mymagicpower)                          |
+| 通用工具 | [Lmdb Dataset Format Conversion Tool](https://github.com/OneYearIsEnough/PaddleOCR-Recog-LmdbDataset-Conversion) | 文本识别任务中lmdb数据格式转换工具 | [OneYearIsEnough](https://github.com/OneYearIsEnough) |
+| 垂类工具 | [AI Studio项目](https://aistudio.baidu.com/aistudio/projectdetail/1054614?channelType=0&channel=0) | 英文视频自动生成字幕       | [叶月水狐](https://aistudio.baidu.com/aistudio/personalcenter/thirdview/322052) |
+| 垂类工具 | [id_card_ocr](https://github.com/baseli/id_card_ocr)         | 身份证复印件识别           | [baseli](https://github.com/baseli)                          |
+| 垂类工具 | [Paddle_Table_Image_Reader](https://github.com/thunder95/Paddle_Table_Image_Reader) | 能看懂表格图片的数据助手   | [thunder95](https://github.com/thunder95])                   |
+| 垂类工具 | [AI Studio项目](https://aistudio.baidu.com/aistudio/projectdetail/3382897) | OCR流程中对手写体进行过滤       | [daassh](https://github.com/daassh) |
+| 垂类工具 | [AI Studio项目](https://aistudio.baidu.com/aistudio/projectdetail/2803693) | 电表读数和编号识别      | [深渊上的坑](https://github.com/edencfc) |
+| 前后处理 | [paddleOCRCorrectOutputs](https://github.com/yuranusduke/paddleOCRCorrectOutputs) | 获取OCR识别结果的key-value | [yuranusduke](https://github.com/yuranusduke)                |
+|前处理| [optlab](https://github.com/GreatV/optlab) |OCR前处理工具箱，基于Qt和Leptonica。|[GreatV](https://github.com/GreatV)|
+|应用部署| [PaddleOCRSharp](https://github.com/raoyutian/PaddleOCRSharp) |PaddleOCR的.NET封装与应用部署。|[raoyutian](https://github.com/raoyutian/PaddleOCRSharp)|
+|应用部署| [PaddleSharp](https://github.com/sdcb/PaddleSharp) |PaddleOCR的.NET封装与应用部署，支持跨平台、GPU|[sdcb](https://github.com/sdcb)|
+| 学术前沿模型训练与推理 | [AI Studio项目](https://aistudio.baidu.com/aistudio/projectdetail/3397137) | StarNet-MobileNetV3算法–中文训练 | [xiaoyangyang2](https://github.com/xiaoyangyang2)                |
 
 ### 1.2 为PaddleOCR新增功能
 

doc/doc_en/models_list_en.md

@@ -43,8 +43,8 @@ Relationship of the above models is as follows.
 
 |model name|description|config|model size|download|
 | --- | --- | --- | --- | --- |
-|ch_PP-OCRv2_rec_slim|[New] Slim qunatization with distillation lightweight model, supporting Chinese, English, multilingual text detection|[ch_PP-OCRv2_rec.yml](../../configs/rec/ch_PP-OCRv2/ch_PP-OCRv2_rec.yml)| 9M |[inference model](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_slim_quant_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_slim_quant_train.tar) |
-|ch_PP-OCRv2_rec|[New] Original lightweight model, supporting Chinese, English, multilingual text detection|[ch_PP-OCRv2_rec.yml](../../configs/rec/ch_PP-OCRv2/ch_PP-OCRv2_rec.yml)|8.5M|[inference model](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_train.tar) |
+|ch_PP-OCRv2_rec_slim|[New] Slim qunatization with distillation lightweight model, supporting Chinese, English, multilingual text recognition|[ch_PP-OCRv2_rec.yml](../../configs/rec/ch_PP-OCRv2/ch_PP-OCRv2_rec.yml)| 9M |[inference model](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_slim_quant_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_slim_quant_train.tar) |
+|ch_PP-OCRv2_rec|[New] Original lightweight model, supporting Chinese, English, multilingual text recognition|[ch_PP-OCRv2_rec_distillation.yml](../../configs/rec/ch_PP-OCRv2/ch_PP-OCRv2_rec_distillation.yml)|8.5M|[inference model](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_train.tar) |
 |ch_ppocr_mobile_slim_v2.0_rec|Slim pruned and quantized lightweight model, supporting Chinese, English and number recognition|[rec_chinese_lite_train_v2.0.yml](../../configs/rec/ch_ppocr_v2.0/rec_chinese_lite_train_v2.0.yml)| 6M | [inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_slim_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_slim_train.tar) |
 |ch_ppocr_mobile_v2.0_rec|Original lightweight model, supporting Chinese, English and number recognition|[rec_chinese_lite_train_v2.0.yml](../../configs/rec/ch_ppocr_v2.0/rec_chinese_lite_train_v2.0.yml)|5.2M|[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_train.tar) / [pre-trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_pre.tar) |
 |ch_ppocr_server_v2.0_rec|General model, supporting Chinese, English and number recognition|[rec_chinese_common_train_v2.0.yml](../../configs/rec/ch_ppocr_v2.0/rec_chinese_common_train_v2.0.yml)|94.8M|[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_train.tar) / [pre-trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_pre.tar) |

notebook/notebook_ch/2.text_detection/文本检测FAQ.ipynb

@@ -81,7 +81,7 @@
     "\n",
     "如果对某些层使用更小的学习率学习，静态图里还不是很方便，一个方法是在参数初始化的时候，给权重的属性设置固定的学习率，参考：https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/fluid/param_attr/ParamAttr_cn.html#paramattr\n",
     "\n",
-    "实际上我们实验发现，直接加载模型去fine-tune，不设置某些层不同学习率，效果也都不错\n",
+    "实际上我们实验发现，直接加载模型去fine-tune，不设置某些层不同学习率，效果也都不错。\n",
     "\n",
     "**1.11 DB的预处理部分，图片的长和宽为什么要处理成32的倍数？**\n",
     "\n",
@@ -95,7 +95,7 @@
     "\n",
     "**1.13 PP-OCR检测效果不好，该如何优化？**\n",
     "\n",
-    "A： 具体问题具体分析:\n",
+    "**A**： 具体问题具体分析:\n",
     "- 如果在你的场景上检测效果不可用，首选是在你的数据上做finetune训练；\n",
     "- 如果图像过大，文字过于密集，建议不要过度压缩图像，可以尝试修改检测预处理的resize逻辑，防止图像被过度压缩；\n",
     "- 检测框大小过于紧贴文字或检测框过大，可以调整db_unclip_ratio这个参数，加大参数可以扩大检测框，减小参数可以减小检测框大小；\n",
@@ -123,8 +123,8 @@
     "\n",
     "**A**：GPU加速预测推荐使用TensorRT。\n",
     "- 1. 从[链接](https://paddleinference.paddlepaddle.org.cn/master/user_guides/download_lib.html)下载带TensorRT的Paddle安装包或者预测库。\n",
-    "- 2. 从Nvidia官网下载TensorRT版本，注意下载的TensorRT版本与paddle安装包中编译的TensorRT版本一致。\n",
-    "- 3. 设置环境变量LD_LIBRARY_PATH，指向TensorRT的lib文件夹\n",
+    "- 2. 从Nvidia官网下载[TensorRT](https://developer.nvidia.com/tensorrt)，注意下载的TensorRT版本与paddle安装包中编译的TensorRT版本一致。\n",
+    "- 3. 设置环境变量`LD_LIBRARY_PATH`，指向TensorRT的lib文件夹\n",
     "```\n",
     "export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:<TensorRT-${version}/lib>\n",
     "```\n",

notebook/notebook_ch/2.text_detection/文本检测理论篇.ipynb

@@ -6,7 +6,7 @@
     "collapsed": false
    },
    "source": [
-    "# OCR七日课之文本检测综述\n"
+    "# 文本检测算法理论\n"
    ]
   },
   {
@@ -15,11 +15,11 @@
     "collapsed": false
    },
    "source": [
-    "## 1. 文本检测\n",
+    "## 1 文本检测\n",
     "\n",
     "文本检测任务是找出图像或视频中的文字位置。不同于目标检测任务，目标检测不仅要解决定位问题，还要解决目标分类问题。\n",
     "\n",
-    "文本在图像中的表现形式可以视为一种‘目标‘，通用的目标检测的方法也适用于文本检测，从任务本身上来看：\n",
+    "文本在图像中的表现形式可以视为一种‘目标’，通用的目标检测的方法也适用于文本检测，从任务本身上来看：\n",
     "\n",
     "- 目标检测：给定图像或者视频，找出目标的位置（box），并给出目标的类别；\n",
     "- 文本检测：给定输入图像或者视频，找出文本的区域，可以是单字符位置或者整个文本行位置；\n",
@@ -41,14 +41,14 @@
     "1. 自然场景中文本具有多样性：文本检测受到文字颜色、大小、字体、形状、方向、语言、以及文本长度的影响；\n",
     "2. 复杂的背景和干扰；文本检测受到图像失真，模糊，低分辨率，阴影，亮度等因素的影响；\n",
     "3. 文本密集甚至重叠会影响文字的检测；\n",
-    "4. 文字存在局部一致性，文本行的一小部分，也可视为是独立的文本；\n",
+    "4. 文字存在局部一致性：文本行的一小部分，也可视为是独立的文本。\n",
     "\n",
     "<center><img src=\"https://ai-studio-static-online.cdn.bcebos.com/072f208f2aff47e886cf2cf1378e23c648356686cf1349c799b42f662d8ced00\"\n",
     "width=\"1000\" ></center>\n",
     "\n",
     "<br><center>图3 文本检测场景</center>\n",
     "\n",
-    "针对以上问题，衍生了很多基于深度学习的文本检测算法，解决自然场景文字检测问题，这些方法可以分为基于回归和基于分割的文本检测方法。\n",
+    "针对以上问题，衍生出了很多基于深度学习的文本检测算法，用于解决自然场景文字检测问题。这些方法可以分为基于回归和基于分割的文本检测方法。\n",
     "\n",
     "下一节将简要介绍基于深度学习技术的经典文字检测算法。"
    ]
@@ -59,7 +59,7 @@
     "collapsed": false
    },
    "source": [
-    "## 2. 文本检测方法介绍\n",
+    "## 2 文本检测方法介绍\n",
     "\n",
     "\n",
     "近些年来基于深度学习的文本检测算法层出不穷，这些方法大致可以分为两类：\n",
@@ -134,7 +134,7 @@
     "\n",
     "\n",
     "\n",
-    "LOMO[19]针对长文本和弯曲文本问题，提出迭代的优化文本定位特征获取更精细的文本定位，该方法包括三个部分，坐标回归模块DR，迭代优化模块IRM以及任意形状表达模块SEM。分别用于生成文本大致区域，迭代优化文本定位特征，预测文本区域、文本中心线以及文本边界。迭代的优化文本特征可以更好的解决长文本定位问题以及获得更精确的文本区域定位。\n",
+    "LOMO[19]针对长文本和弯曲文本问题，提出迭代的优化文本定位特征获取更精细的文本定位。该方法包括三个部分：坐标回归模块DR，迭代优化模块IRM以及任意形状表达模块SEM。它们分别用于生成文本大致区域，迭代优化文本定位特征，预测文本区域、文本中心线以及文本边界。迭代的优化文本特征可以更好的解决长文本定位问题以及获得更精确的文本区域定位。\n",
     "<center><img src=\"https://ai-studio-static-online.cdn.bcebos.com/e90adf3ca25a45a0af0b84a181fbe2c4954be1fcca8f4049957128548b7131ef\"\n",
     "width=\"1000\" ></center>\n",
     "<br><center>图11 LOMO框架图</center>\n",
@@ -228,7 +228,7 @@
     "collapsed": false
    },
    "source": [
-    "## 3. 总结\n",
+    "## 3 总结\n",
     "\n",
     "本节介绍了近几年来文本检测领域的发展，包括基于回归、分割的文本检测方法，并分别列举并介绍了一些经典论文的方法思路。下一节以PaddleOCR开源库为例，详细介绍DBNet的算法原理以及核心代码实现。"
    ]

notebook/notebook_ch/如何使用本书.ipynb

@@ -42,7 +42,7 @@
     "\n",
     "然后安装第三方库：\n",
     "\n",
-    "```\n",
+    "```bash\n",
     "cd PaddleOCR\n",
     "pip3 install -r requirements.txt\n",
     "```\n",