Skip to content

GitLab

"vscode:/vscode.git/clone" did not exist on "c2a7e7f5101c7f2248a60152eb0c647abe2ed983"
Unverified Commit 3df64d50 authored by zhoujun's avatar zhoujun Committed by GitHub
Browse files

Merge pull request #5576 from WenmuZhou/android 

[android demo] Fix error when run same image
parents 816be4d7 056b7606
Hide whitespace changes
Inline Side-by-side
Showing with 195 additions and 296 deletions
deploy/android_demo/app/src/main/java/com/baidu/paddle/lite/demo/ocr/Predictor.java
View file @ 3df64d50
......@@ -31,23 +31,19 @@ public class Predictor {
protected float inferenceTime = 0;
// Only for object detection
protected Vector<String> wordLabels = new Vector<String>();
protected String inputColorFormat = "BGR";
protected long[] inputShape = new long[]{1, 3, 960};
protected float[] inputMean = new float[]{0.485f, 0.456f, 0.406f};
protected float[] inputStd = new float[]{1.0f / 0.229f, 1.0f / 0.224f, 1.0f / 0.225f};
protected int detLongSize = 960;
protected float scoreThreshold = 0.1f;
protected Bitmap inputImage = null;
protected Bitmap outputImage = null;
protected volatile String outputResult = "";
protected float preprocessTime = 0;
protected float postprocessTime = 0;
public Predictor() {
}
public boolean init(Context appCtx, String modelPath, String labelPath) {
isLoaded = loadModel(appCtx, modelPath, cpuThreadNum, cpuPowerMode);
public boolean init(Context appCtx, String modelPath, String labelPath, int useOpencl, int cpuThreadNum, String cpuPowerMode) {
isLoaded = loadModel(appCtx, modelPath, useOpencl, cpuThreadNum, cpuPowerMode);
if (!isLoaded) {
return false;
}
......@@ -56,49 +52,18 @@ public class Predictor {
}
public boolean init(Context appCtx, String modelPath, String labelPath, int cpuThreadNum, String cpuPowerMode,
String inputColorFormat,
long[] inputShape, float[] inputMean,
float[] inputStd, float scoreThreshold) {
if (inputShape.length != 3) {
Log.e(TAG, "Size of input shape should be: 3");
return false;
}
if (inputMean.length != inputShape[1]) {
Log.e(TAG, "Size of input mean should be: " + Long.toString(inputShape[1]));
return false;
}
if (inputStd.length != inputShape[1]) {
Log.e(TAG, "Size of input std should be: " + Long.toString(inputShape[1]));
return false;
}
if (inputShape[0] != 1) {
Log.e(TAG, "Only one batch is supported in the image classification demo, you can use any batch size in " +
"your Apps!");
return false;
}
if (inputShape[1] != 1 && inputShape[1] != 3) {
Log.e(TAG, "Only one/three channels are supported in the image classification demo, you can use any " +
"channel size in your Apps!");
return false;
}
if (!inputColorFormat.equalsIgnoreCase("BGR")) {
Log.e(TAG, "Only BGR color format is supported.");
return false;
}
boolean isLoaded = init(appCtx, modelPath, labelPath);
public boolean init(Context appCtx, String modelPath, String labelPath, int useOpencl, int cpuThreadNum, String cpuPowerMode,
int detLongSize, float scoreThreshold) {
boolean isLoaded = init(appCtx, modelPath, labelPath, useOpencl, cpuThreadNum, cpuPowerMode);
if (!isLoaded) {
return false;
}
this.inputColorFormat = inputColorFormat;
this.inputShape = inputShape;
this.inputMean = inputMean;
this.inputStd = inputStd;
this.detLongSize = detLongSize;
this.scoreThreshold = scoreThreshold;
return true;
}
protected boolean loadModel(Context appCtx, String modelPath, int cpuThreadNum, String cpuPowerMode) {
protected boolean loadModel(Context appCtx, String modelPath, int useOpencl, int cpuThreadNum, String cpuPowerMode) {
// Release model if exists
releaseModel();
......@@ -118,12 +83,13 @@ public class Predictor {
}
OCRPredictorNative.Config config = new OCRPredictorNative.Config();
config.useOpencl = useOpencl;
config.cpuThreadNum = cpuThreadNum;
config.detModelFilename = realPath + File.separator + "ch_ppocr_mobile_v2.0_det_opt.nb";
config.recModelFilename = realPath + File.separator + "ch_ppocr_mobile_v2.0_rec_opt.nb";
config.clsModelFilename = realPath + File.separator + "ch_ppocr_mobile_v2.0_cls_opt.nb";
Log.e("Predictor", "model path" + config.detModelFilename + " ; " + config.recModelFilename + ";" + config.clsModelFilename);
config.cpuPower = cpuPowerMode;
config.detModelFilename = realPath + File.separator + "det_db.nb";
config.recModelFilename = realPath + File.separator + "rec_crnn.nb";
config.clsModelFilename = realPath + File.separator + "cls.nb";
Log.i("Predictor", "model path" + config.detModelFilename + " ; " + config.recModelFilename + ";" + config.clsModelFilename);
paddlePredictor = new OCRPredictorNative(config);
this.cpuThreadNum = cpuThreadNum;
......@@ -170,82 +136,29 @@ public class Predictor {
}
public boolean runModel() {
public boolean runModel(int run_det, int run_cls, int run_rec) {
if (inputImage == null || !isLoaded()) {
return false;
}
// Pre-process image, and feed input tensor with pre-processed data
Bitmap scaleImage = Utils.resizeWithStep(inputImage, Long.valueOf(inputShape[2]).intValue(), 32);
Date start = new Date();
int channels = (int) inputShape[1];
int width = scaleImage.getWidth();
int height = scaleImage.getHeight();
float[] inputData = new float[channels * width * height];
if (channels == 3) {
int[] channelIdx = null;
if (inputColorFormat.equalsIgnoreCase("RGB")) {
channelIdx = new int[]{0, 1, 2};
} else if (inputColorFormat.equalsIgnoreCase("BGR")) {
channelIdx = new int[]{2, 1, 0};
} else {
Log.i(TAG, "Unknown color format " + inputColorFormat + ", only RGB and BGR color format is " +
"supported!");
return false;
}
int[] channelStride = new int[]{width * height, width * height * 2};
int[] pixels=new int[width*height];
scaleImage.getPixels(pixels,0,scaleImage.getWidth(),0,0,scaleImage.getWidth(),scaleImage.getHeight());
for (int i = 0; i < pixels.length; i++) {
int color = pixels[i];
float[] rgb = new float[]{(float) red(color) / 255.0f, (float) green(color) / 255.0f,
(float) blue(color) / 255.0f};
inputData[i] = (rgb[channelIdx[0]] - inputMean[0]) / inputStd[0];
inputData[i + channelStride[0]] = (rgb[channelIdx[1]] - inputMean[1]) / inputStd[1];
inputData[i+ channelStride[1]] = (rgb[channelIdx[2]] - inputMean[2]) / inputStd[2];
}
} else if (channels == 1) {
int[] pixels=new int[width*height];
scaleImage.getPixels(pixels,0,scaleImage.getWidth(),0,0,scaleImage.getWidth(),scaleImage.getHeight());
for (int i = 0; i < pixels.length; i++) {
int color = pixels[i];
float gray = (float) (red(color) + green(color) + blue(color)) / 3.0f / 255.0f;
inputData[i] = (gray - inputMean[0]) / inputStd[0];
}
} else {
Log.i(TAG, "Unsupported channel size " + Integer.toString(channels) + ", only channel 1 and 3 is " +
"supported!");
return false;
}
float[] pixels = inputData;
Log.i(TAG, "pixels " + pixels[0] + " " + pixels[1] + " " + pixels[2] + " " + pixels[3]
+ " " + pixels[pixels.length / 2] + " " + pixels[pixels.length / 2 + 1] + " " + pixels[pixels.length - 2] + " " + pixels[pixels.length - 1]);
Date end = new Date();
preprocessTime = (float) (end.getTime() - start.getTime());
// Warm up
for (int i = 0; i < warmupIterNum; i++) {
paddlePredictor.runImage(inputData, width, height, channels, inputImage);
paddlePredictor.runImage(inputImage, detLongSize, run_det, run_cls, run_rec);
}
warmupIterNum = 0; // do not need warm
// Run inference
start = new Date();
ArrayList<OcrResultModel> results = paddlePredictor.runImage(inputData, width, height, channels, inputImage);
end = new Date();
Date start = new Date();
ArrayList<OcrResultModel> results = paddlePredictor.runImage(inputImage, detLongSize, run_det, run_cls, run_rec);
Date end = new Date();
inferenceTime = (end.getTime() - start.getTime()) / (float) inferIterNum;
results = postprocess(results);
Log.i(TAG, "[stat] Preprocess Time: " + preprocessTime
+ " ; Inference Time: " + inferenceTime + " ;Box Size " + results.size());
Log.i(TAG, "[stat] Inference Time: " + inferenceTime + " ;Box Size " + results.size());
drawResults(results);
return true;
}
public boolean isLoaded() {
return paddlePredictor != null && isLoaded;
}
......@@ -282,10 +195,6 @@ public class Predictor {
return outputResult;
}
public float preprocessTime() {
return preprocessTime;
}
public float postprocessTime() {
return postprocessTime;
}
......@@ -310,6 +219,7 @@ public class Predictor {
}
}
r.setLabel(word.toString());
r.setClsLabel(r.getClsIdx() == 1 ? "180" : "0");
}
return results;
}
......@@ -319,14 +229,22 @@ public class Predictor {
for (int i = 0; i < results.size(); i++) {
OcrResultModel result = results.get(i);
StringBuilder sb = new StringBuilder("");
sb.append(result.getLabel());
sb.append(" ").append(result.getConfidence());
sb.append("; Points: ");
for (Point p : result.getPoints()) {
sb.append("(").append(p.x).append(",").append(p.y).append(") ");
if(result.getPoints().size()>0){
sb.append("Det: ");
for (Point p : result.getPoints()) {
sb.append("(").append(p.x).append(",").append(p.y).append(") ");
}
}
if(result.getLabel().length() > 0){
sb.append("\n Rec: ").append(result.getLabel());
sb.append(",").append(result.getConfidence());
}
if(result.getClsIdx()!=-1){
sb.append(" Cls: ").append(result.getClsLabel());
sb.append(",").append(result.getClsConfidence());
}
Log.i(TAG, sb.toString()); // show LOG in Logcat panel
outputResultSb.append(i + 1).append(": ").append(result.getLabel()).append("\n");
outputResultSb.append(i + 1).append(": ").append(sb.toString()).append("\n");
}
outputResult = outputResultSb.toString();
outputImage = inputImage;
......@@ -344,6 +262,9 @@ public class Predictor {
for (OcrResultModel result : results) {
Path path = new Path();
List<Point> points = result.getPoints();
if(points.size()==0){
continue;
}
path.moveTo(points.get(0).x, points.get(0).y);
for (int i = points.size() - 1; i >= 0; i--) {
Point p = points.get(i);
......
deploy/android_demo/app/src/main/java/com/baidu/paddle/lite/demo/ocr/SettingsActivity.java
View file @ 3df64d50
......@@ -20,16 +20,13 @@ public class SettingsActivity extends AppCompatPreferenceActivity implements Sha
ListPreference etImagePath = null;
ListPreference lpCPUThreadNum = null;
ListPreference lpCPUPowerMode = null;
ListPreference lpInputColorFormat = null;
EditTextPreference etInputShape = null;
EditTextPreference etInputMean = null;
EditTextPreference etInputStd = null;
EditTextPreference etDetLongSize = null;
EditTextPreference etScoreThreshold = null;
List<String> preInstalledModelPaths = null;
List<String> preInstalledLabelPaths = null;
List<String> preInstalledImagePaths = null;
List<String> preInstalledInputShapes = null;
List<String> preInstalledDetLongSizes = null;
List<String> preInstalledCPUThreadNums = null;
List<String> preInstalledCPUPowerModes = null;
List<String> preInstalledInputColorFormats = null;
......@@ -50,7 +47,7 @@ public class SettingsActivity extends AppCompatPreferenceActivity implements Sha
preInstalledModelPaths = new ArrayList<String>();
preInstalledLabelPaths = new ArrayList<String>();
preInstalledImagePaths = new ArrayList<String>();
preInstalledInputShapes = new ArrayList<String>();
preInstalledDetLongSizes = new ArrayList<String>();
preInstalledCPUThreadNums = new ArrayList<String>();
preInstalledCPUPowerModes = new ArrayList<String>();
preInstalledInputColorFormats = new ArrayList<String>();
......@@ -63,10 +60,7 @@ public class SettingsActivity extends AppCompatPreferenceActivity implements Sha
preInstalledImagePaths.add(getString(R.string.IMAGE_PATH_DEFAULT));
preInstalledCPUThreadNums.add(getString(R.string.CPU_THREAD_NUM_DEFAULT));
preInstalledCPUPowerModes.add(getString(R.string.CPU_POWER_MODE_DEFAULT));
preInstalledInputColorFormats.add(getString(R.string.INPUT_COLOR_FORMAT_DEFAULT));
preInstalledInputShapes.add(getString(R.string.INPUT_SHAPE_DEFAULT));
preInstalledInputMeans.add(getString(R.string.INPUT_MEAN_DEFAULT));
preInstalledInputStds.add(getString(R.string.INPUT_STD_DEFAULT));
preInstalledDetLongSizes.add(getString(R.string.DET_LONG_SIZE_DEFAULT));
preInstalledScoreThresholds.add(getString(R.string.SCORE_THRESHOLD_DEFAULT));
// Setup UI components
......@@ -89,11 +83,7 @@ public class SettingsActivity extends AppCompatPreferenceActivity implements Sha
(ListPreference) findPreference(getString(R.string.CPU_THREAD_NUM_KEY));
lpCPUPowerMode =
(ListPreference) findPreference(getString(R.string.CPU_POWER_MODE_KEY));
lpInputColorFormat =
(ListPreference) findPreference(getString(R.string.INPUT_COLOR_FORMAT_KEY));
etInputShape = (EditTextPreference) findPreference(getString(R.string.INPUT_SHAPE_KEY));
etInputMean = (EditTextPreference) findPreference(getString(R.string.INPUT_MEAN_KEY));
etInputStd = (EditTextPreference) findPreference(getString(R.string.INPUT_STD_KEY));
etDetLongSize = (EditTextPreference) findPreference(getString(R.string.DET_LONG_SIZE_KEY));
etScoreThreshold = (EditTextPreference) findPreference(getString(R.string.SCORE_THRESHOLD_KEY));
}
......@@ -112,11 +102,7 @@ public class SettingsActivity extends AppCompatPreferenceActivity implements Sha
editor.putString(getString(R.string.IMAGE_PATH_KEY), preInstalledImagePaths.get(modelIdx));
editor.putString(getString(R.string.CPU_THREAD_NUM_KEY), preInstalledCPUThreadNums.get(modelIdx));
editor.putString(getString(R.string.CPU_POWER_MODE_KEY), preInstalledCPUPowerModes.get(modelIdx));
editor.putString(getString(R.string.INPUT_COLOR_FORMAT_KEY),
preInstalledInputColorFormats.get(modelIdx));
editor.putString(getString(R.string.INPUT_SHAPE_KEY), preInstalledInputShapes.get(modelIdx));
editor.putString(getString(R.string.INPUT_MEAN_KEY), preInstalledInputMeans.get(modelIdx));
editor.putString(getString(R.string.INPUT_STD_KEY), preInstalledInputStds.get(modelIdx));
editor.putString(getString(R.string.DET_LONG_SIZE_KEY), preInstalledDetLongSizes.get(modelIdx));
editor.putString(getString(R.string.SCORE_THRESHOLD_KEY),
preInstalledScoreThresholds.get(modelIdx));
editor.apply();
......@@ -129,10 +115,7 @@ public class SettingsActivity extends AppCompatPreferenceActivity implements Sha
etImagePath.setEnabled(enableCustomSettings);
lpCPUThreadNum.setEnabled(enableCustomSettings);
lpCPUPowerMode.setEnabled(enableCustomSettings);
lpInputColorFormat.setEnabled(enableCustomSettings);
etInputShape.setEnabled(enableCustomSettings);
etInputMean.setEnabled(enableCustomSettings);
etInputStd.setEnabled(enableCustomSettings);
etDetLongSize.setEnabled(enableCustomSettings);
etScoreThreshold.setEnabled(enableCustomSettings);
modelPath = sharedPreferences.getString(getString(R.string.MODEL_PATH_KEY),
getString(R.string.MODEL_PATH_DEFAULT));
......@@ -144,14 +127,8 @@ public class SettingsActivity extends AppCompatPreferenceActivity implements Sha
getString(R.string.CPU_THREAD_NUM_DEFAULT));
String cpuPowerMode = sharedPreferences.getString(getString(R.string.CPU_POWER_MODE_KEY),
getString(R.string.CPU_POWER_MODE_DEFAULT));
String inputColorFormat = sharedPreferences.getString(getString(R.string.INPUT_COLOR_FORMAT_KEY),
getString(R.string.INPUT_COLOR_FORMAT_DEFAULT));
String inputShape = sharedPreferences.getString(getString(R.string.INPUT_SHAPE_KEY),
getString(R.string.INPUT_SHAPE_DEFAULT));
String inputMean = sharedPreferences.getString(getString(R.string.INPUT_MEAN_KEY),
getString(R.string.INPUT_MEAN_DEFAULT));
String inputStd = sharedPreferences.getString(getString(R.string.INPUT_STD_KEY),
getString(R.string.INPUT_STD_DEFAULT));
String detLongSize = sharedPreferences.getString(getString(R.string.DET_LONG_SIZE_KEY),
getString(R.string.DET_LONG_SIZE_DEFAULT));
String scoreThreshold = sharedPreferences.getString(getString(R.string.SCORE_THRESHOLD_KEY),
getString(R.string.SCORE_THRESHOLD_DEFAULT));
etModelPath.setSummary(modelPath);
......@@ -164,14 +141,8 @@ public class SettingsActivity extends AppCompatPreferenceActivity implements Sha
lpCPUThreadNum.setSummary(cpuThreadNum);
lpCPUPowerMode.setValue(cpuPowerMode);
lpCPUPowerMode.setSummary(cpuPowerMode);
lpInputColorFormat.setValue(inputColorFormat);
lpInputColorFormat.setSummary(inputColorFormat);
etInputShape.setSummary(inputShape);
etInputShape.setText(inputShape);
etInputMean.setSummary(inputMean);
etInputMean.setText(inputMean);
etInputStd.setSummary(inputStd);
etInputStd.setText(inputStd);
etDetLongSize.setSummary(detLongSize);
etDetLongSize.setText(detLongSize);
etScoreThreshold.setText(scoreThreshold);
etScoreThreshold.setSummary(scoreThreshold);
}
......
deploy/android_demo/app/src/main/res/layout/activity_main.xml
View file @ 3df64d50
......@@ -23,13 +23,7 @@
android:layout_height="wrap_content"
android:orientation="horizontal">
<Button
android:id="@+id/btn_load_model"
android:layout_width="0dp"
android:layout_height="wrap_content"
android:layout_weight="1"
android:onClick="btn_load_model_click"
android:text="加载模型" />
<Button
android:id="@+id/btn_run_model"
android:layout_width="0dp"
......@@ -52,7 +46,45 @@
android:onClick="btn_choice_img_click"
android:text="选取图片" />
<Button
android:id="@+id/btn_reset_img"
android:layout_width="0dp"
android:layout_height="wrap_content"
android:layout_weight="1"
android:onClick="btn_reset_img_click"
android:text="清空绘图" />
</LinearLayout>
<LinearLayout
android:id="@+id/run_mode_layout"
android:layout_width="fill_parent"
android:layout_height="wrap_content"
android:orientation="horizontal">
<CheckBox
android:id="@+id/cb_opencl"
android:layout_width="0dp"
android:layout_weight="1"
android:layout_height="wrap_content"
android:text="开启OPENCL"
android:onClick="cb_opencl_click"
android:visibility="gone"/>
<TextView
android:layout_width="0dp"
android:layout_weight="0.5"
android:layout_height="wrap_content"
android:text="运行模式:"/>
<Spinner
android:id="@+id/sp_run_mode"
android:layout_width="0dp"
android:layout_weight="1.5"
android:layout_height="wrap_content"
android:entries="@array/run_Model"
/>
</LinearLayout>
<TextView
android:id="@+id/tv_input_setting"
android:layout_width="wrap_content"
......@@ -60,7 +92,7 @@
android:scrollbars="vertical"
android:layout_marginLeft="12dp"
android:layout_marginRight="12dp"
android:layout_marginTop="10dp"
android:layout_marginTop="5dp"
android:layout_marginBottom="5dp"
android:lineSpacingExtra="4dp"
android:singleLine="false"
......
deploy/android_demo/app/src/main/res/layout/activity_mini.xml deleted 100644 → 0
View file @ 816be4d7
<?xml version="1.0" encoding="utf-8"?>
<!-- for MiniActivity Use Only -->
<androidx.constraintlayout.widget.ConstraintLayout xmlns:android="http://schemas.android.com/apk/res/android"
xmlns:app="http://schemas.android.com/apk/res-auto"
xmlns:tools="http://schemas.android.com/tools"
android:layout_width="match_parent"
android:layout_height="match_parent"
app:layout_constraintLeft_toLeftOf="parent"
app:layout_constraintLeft_toRightOf="parent"
tools:context=".MainActivity">
<TextView
android:id="@+id/sample_text"
android:layout_width="0dp"
android:layout_height="wrap_content"
android:text="Hello World!"
app:layout_constraintLeft_toLeftOf="parent"
app:layout_constraintRight_toRightOf="parent"
app:layout_constraintTop_toBottomOf="@id/imageView"
android:scrollbars="vertical"
/>
<ImageView
android:id="@+id/imageView"
android:layout_width="wrap_content"
android:layout_height="wrap_content"
android:paddingTop="20dp"
android:paddingBottom="20dp"
app:layout_constraintBottom_toTopOf="@id/imageView"
app:layout_constraintLeft_toLeftOf="parent"
app:layout_constraintRight_toRightOf="parent"
app:layout_constraintTop_toTopOf="parent"
tools:srcCompat="@tools:sample/avatars" />
<Button
android:id="@+id/button"
android:layout_width="wrap_content"
android:layout_height="wrap_content"
android:layout_marginBottom="4dp"
android:text="Button"
app:layout_constraintBottom_toBottomOf="parent"
app:layout_constraintLeft_toLeftOf="parent"
app:layout_constraintRight_toRightOf="parent"
tools:layout_editor_absoluteX="161dp" />
</androidx.constraintlayout.widget.ConstraintLayout>
\ No newline at end of file
deploy/android_demo/app/src/main/res/values/arrays.xml
View file @ 3df64d50
<?xml version="1.0" encoding="utf-8"?>
<resources>
<string-array name="image_name_entries">
<item>0.jpg</item>
<item>90.jpg</item>
<item>180.jpg</item>
<item>270.jpg</item>
<item>det_0.jpg</item>
<item>det_90.jpg</item>
<item>det_180.jpg</item>
<item>det_270.jpg</item>
<item>rec_0.jpg</item>
<item>rec_0_180.jpg</item>
<item>rec_1.jpg</item>
<item>rec_1_180.jpg</item>
</string-array>
<string-array name="image_name_values">
<item>images/0.jpg</item>
<item>images/90.jpg</item>
<item>images/180.jpg</item>
<item>images/270.jpg</item>
<item>images/det_0.jpg</item>
<item>images/det_90.jpg</item>
<item>images/det_180.jpg</item>
<item>images/det_270.jpg</item>
<item>images/rec_0.jpg</item>
<item>images/rec_0_180.jpg</item>
<item>images/rec_1.jpg</item>
<item>images/rec_1_180.jpg</item>
</string-array>
<string-array name="cpu_thread_num_entries">
<item>1 threads</item>
......@@ -48,4 +56,12 @@
<item>BGR</item>
<item>RGB</item>
</string-array>
<string-array name="run_Model">
<item>检测+分类+识别</item>
<item>检测+识别</item>
<item>分类+识别</item>
<item>检测</item>
<item>识别</item>
<item>分类</item>
</string-array>
</resources>
\ No newline at end of file
deploy/android_demo/app/src/main/res/values/strings.xml
View file @ 3df64d50
<resources>
<string name="app_name">OCR Chinese</string>
<string name="app_name">PaddleOCR</string>
<string name="CHOOSE_PRE_INSTALLED_MODEL_KEY">CHOOSE_PRE_INSTALLED_MODEL_KEY</string>
<string name="ENABLE_CUSTOM_SETTINGS_KEY">ENABLE_CUSTOM_SETTINGS_KEY</string>
<string name="MODEL_PATH_KEY">MODEL_PATH_KEY</string>
......@@ -7,20 +7,14 @@
<string name="IMAGE_PATH_KEY">IMAGE_PATH_KEY</string>
<string name="CPU_THREAD_NUM_KEY">CPU_THREAD_NUM_KEY</string>
<string name="CPU_POWER_MODE_KEY">CPU_POWER_MODE_KEY</string>
<string name="INPUT_COLOR_FORMAT_KEY">INPUT_COLOR_FORMAT_KEY</string>
<string name="INPUT_SHAPE_KEY">INPUT_SHAPE_KEY</string>
<string name="INPUT_MEAN_KEY">INPUT_MEAN_KEY</string>
<string name="INPUT_STD_KEY">INPUT_STD_KEY</string>
<string name="DET_LONG_SIZE_KEY">DET_LONG_SIZE_KEY</string>
<string name="SCORE_THRESHOLD_KEY">SCORE_THRESHOLD_KEY</string>
<string name="MODEL_PATH_DEFAULT">models/ocr_v2_for_cpu</string>
<string name="MODEL_PATH_DEFAULT">models/ch_PP-OCRv2</string>
<string name="LABEL_PATH_DEFAULT">labels/ppocr_keys_v1.txt</string>
<string name="IMAGE_PATH_DEFAULT">images/0.jpg</string>
<string name="IMAGE_PATH_DEFAULT">images/det_0.jpg</string>
<string name="CPU_THREAD_NUM_DEFAULT">4</string>
<string name="CPU_POWER_MODE_DEFAULT">LITE_POWER_HIGH</string>
<string name="INPUT_COLOR_FORMAT_DEFAULT">BGR</string>
<string name="INPUT_SHAPE_DEFAULT">1,3,960</string>
<string name="INPUT_MEAN_DEFAULT">0.485, 0.456, 0.406</string>
<string name="INPUT_STD_DEFAULT">0.229,0.224,0.225</string>
<string name="DET_LONG_SIZE_DEFAULT">960</string>
<string name="SCORE_THRESHOLD_DEFAULT">0.1</string>
</resources>
deploy/android_demo/app/src/main/res/xml/settings.xml
View file @ 3df64d50
......@@ -47,26 +47,10 @@
android:entryValues="@array/cpu_power_mode_values"/>
</PreferenceCategory>
<PreferenceCategory android:title="Input Settings">
<ListPreference
android:defaultValue="@string/INPUT_COLOR_FORMAT_DEFAULT"
android:key="@string/INPUT_COLOR_FORMAT_KEY"
android:negativeButtonText="@null"
android:positiveButtonText="@null"
android:title="Input Color Format: BGR or RGB"
android:entries="@array/input_color_format_entries"
android:entryValues="@array/input_color_format_values"/>
<EditTextPreference
android:key="@string/INPUT_SHAPE_KEY"
android:defaultValue="@string/INPUT_SHAPE_DEFAULT"
android:title="Input Shape: (1,1,max_width_height) or (1,3,max_width_height)" />
<EditTextPreference
android:key="@string/INPUT_MEAN_KEY"
android:defaultValue="@string/INPUT_MEAN_DEFAULT"
android:title="Input Mean: (channel/255-mean)/std" />
<EditTextPreference
android:key="@string/INPUT_STD_KEY"
android:defaultValue="@string/INPUT_STD_DEFAULT"
android:title="Input Std: (channel/255-mean)/std" />
android:key="@string/DET_LONG_SIZE_KEY"
android:defaultValue="@string/DET_LONG_SIZE_DEFAULT"
android:title="det long size" />
</PreferenceCategory>
<PreferenceCategory android:title="Output Settings">
<EditTextPreference
......
deploy/lite/readme.md
View file @ 3df64d50
- [端侧部署](#端侧部署)
- [1. 准备环境](#1-准备环境)
- [运行准备](#运行准备)
- [1.1 准备交叉编译环境](#11-准备交叉编译环境)
- [1.2 准备预测库](#12-准备预测库)
- [2 开始运行](#2-开始运行)
- [2.1 模型优化](#21-模型优化)
- [2.2 与手机联调](#22-与手机联调)
- [注意：](#注意)
- [FAQ](#faq)
# 端侧部署
本教程将介绍基于[Paddle Lite](https://github.com/PaddlePaddle/Paddle-Lite) 在移动端部署PaddleOCR超轻量中文检测、识别模型的详细步骤。
......@@ -26,17 +37,17 @@ Paddle Lite是飞桨轻量化推理引擎，为手机、IOT端提供高效推理
| 平台 | 预测库下载链接 |
|---|---|
|Android|[arm7](https://github.com/PaddlePaddle/Paddle-Lite/releases/download/v2.9/inference_lite_lib.android.armv7.gcc.c++_shared.with_extra.with_cv.tar.gz) / [arm8](https://github.com/PaddlePaddle/Paddle-Lite/releases/download/v2.9/inference_lite_lib.android.armv8.gcc.c++_shared.with_extra.with_cv.tar.gz)|
|IOS|[arm7](https://github.com/PaddlePaddle/Paddle-Lite/releases/download/v2.9/inference_lite_lib.ios.armv7.with_cv.with_extra.with_log.tiny_publish.tar.gz) / [arm8](https://github.com/PaddlePaddle/Paddle-Lite/releases/download/v2.9/inference_lite_lib.ios.armv8.with_cv.with_extra.with_log.tiny_publish.tar.gz)|
|Android|[arm7](https://github.com/PaddlePaddle/Paddle-Lite/releases/download/v2.10/inference_lite_lib.android.armv7.gcc.c++_shared.with_extra.with_cv.tar.gz) / [arm8](https://github.com/PaddlePaddle/Paddle-Lite/releases/download/v2.10/inference_lite_lib.android.armv8.gcc.c++_shared.with_extra.with_cv.tar.gz)|
|IOS|[arm7](https://github.com/PaddlePaddle/Paddle-Lite/releases/download/v2.10/inference_lite_lib.ios.armv7.with_cv.with_extra.with_log.tiny_publish.tar.gz) / [arm8](https://github.com/PaddlePaddle/Paddle-Lite/releases/download/v2.10/inference_lite_lib.ios.armv8.with_cv.with_extra.with_log.tiny_publish.tar.gz)|
注：1. 上述预测库为PaddleLite 2.9分支编译得到，有关PaddleLite 2.9 详细信息可参考 [链接](https://github.com/PaddlePaddle/Paddle-Lite/releases/tag/v2.9) 。
注：1. 上述预测库为PaddleLite 2.10分支编译得到，有关PaddleLite 2.10 详细信息可参考 [链接](https://github.com/PaddlePaddle/Paddle-Lite/releases/tag/v2.10) 。
- 2. [推荐]编译Paddle-Lite得到预测库，Paddle-Lite的编译方式如下：
```
git clone https://github.com/PaddlePaddle/Paddle-Lite.git
cd Paddle-Lite
# 切换到Paddle-Lite release/v2.9 稳定分支
git checkout release/v2.9
# 切换到Paddle-Lite release/v2.10 稳定分支
git checkout release/v2.10
./lite/tools/build_android.sh --arch=armv8 --with_cv=ON --with_extra=ON
```
......@@ -85,8 +96,8 @@ Paddle-Lite 提供了多种策略来自动优化原始的模型，其中包括
|模型版本|模型简介|模型大小|检测模型|文本方向分类模型|识别模型|Paddle-Lite版本|
|---|---|---|---|---|---|---|
|V2.0|超轻量中文OCR 移动端模型|7.8M|[下载地址](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_det_opt.nb)|[下载地址](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_cls_opt.nb)|[下载地址](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_rec_opt.nb)|v2.9|
|V2.0(slim)|超轻量中文OCR 移动端模型|3.3M|[下载地址](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_det_slim_opt.nb)|[下载地址](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_cls_slim_opt.nb)|[下载地址](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_rec_slim_opt.nb)|v2.9|
|PP-OCRv2|蒸馏版超轻量中文OCR移动端模型|11M|[下载地址](https://paddleocr.bj.bcebos.com/PP-OCRv2/lite/ch_PP-OCRv2_det_infer_opt.nb)|[下载地址](https://paddleocr.bj.bcebos.com/PP-OCRv2/lite/ch_ppocr_mobile_v2.0_cls_infer_opt.nb)|[下载地址](https://paddleocr.bj.bcebos.com/PP-OCRv2/lite/ch_PP-OCRv2_rec_infer_opt.nb)|v2.10|
|PP-OCRv2(slim)|蒸馏版超轻量中文OCR移动端模型|4.6M|[下载地址](https://paddleocr.bj.bcebos.com/PP-OCRv2/lite/ch_PP-OCRv2_det_slim_opt.nb)|[下载地址](https://paddleocr.bj.bcebos.com/PP-OCRv2/lite/ch_ppocr_mobile_v2.0_cls_slim_opt.nb)|[下载地址](https://paddleocr.bj.bcebos.com/PP-OCRv2/lite/ch_PP-OCRv2_rec_slim_opt.nb)|v2.10|
如果直接使用上述表格中的模型进行部署，可略过下述步骤，直接阅读 [2.2节](#2.2与手机联调)。
......@@ -97,7 +108,7 @@ Paddle-Lite 提供了多种策略来自动优化原始的模型，其中包括
# 如果准备环境时已经clone了Paddle-Lite，则不用重新clone Paddle-Lite
git clone https://github.com/PaddlePaddle/Paddle-Lite.git
cd Paddle-Lite
git checkout release/v2.9
git checkout release/v2.10
# 启动编译
./lite/tools/build.sh build_optimize_tool
```
......@@ -123,15 +134,15 @@ cd build.opt/lite/api/
下面以PaddleOCR的超轻量中文模型为例，介绍使用编译好的opt文件完成inference模型到Paddle-Lite优化模型的转换。
```
# 【推荐】 下载PaddleOCR V2.0版本的中英文 inference模型
wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/slim/ch_ppocr_mobile_v2.0_det_slim_infer.tar && tar xf ch_ppocr_mobile_v2.0_det_slim_infer.tar
wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/slim/ch_ppocr_mobile_v2.0_rec_slim_infer.tar && tar xf ch_ppocr_mobile_v2.0_rec_slim_infer.tar
# 【推荐】 下载 PP-OCRv2版本的中英文 inference模型
wget https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_det_slim_quant_infer.tar && tar xf ch_PP-OCRv2_det_slim_quant_infer.tar
wget https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_slim_quant_infer.tar && tar xf ch_PP-OCRv2_rec_slim_quant_infer.tar
wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/slim/ch_ppocr_mobile_v2.0_cls_slim_infer.tar && tar xf ch_ppocr_mobile_v2.0_cls_slim_infer.tar
# 转换V2.0检测模型
./opt --model_file=./ch_ppocr_mobile_v2.0_det_slim_infer/inference.pdmodel --param_file=./ch_ppocr_mobile_v2.0_det_slim_infer/inference.pdiparams --optimize_out=./ch_ppocr_mobile_v2.0_det_slim_opt --valid_targets=arm --optimize_out_type=naive_buffer
# 转换V2.0识别模型
./opt --model_file=./ch_ppocr_mobile_v2.0_rec_slim_infer/inference.pdmodel --param_file=./ch_ppocr_mobile_v2.0_rec_slim_infer/inference.pdiparams --optimize_out=./ch_ppocr_mobile_v2.0_rec_slim_opt --valid_targets=arm --optimize_out_type=naive_buffer
# 转换V2.0方向分类器模型
# 转换检测模型
./opt --model_file=./ch_PP-OCRv2_det_slim_quant_infer/inference.pdmodel --param_file=./ch_PP-OCRv2_det_slim_quant_infer/inference.pdiparams --optimize_out=./ch_PP-OCRv2_det_slim_opt --valid_targets=arm --optimize_out_type=naive_buffer
# 转换识别模型
./opt --model_file=./ch_PP-OCRv2_rec_slim_quant_infer/inference.pdmodel --param_file=./ch_PP-OCRv2_rec_slim_quant_infer/inference.pdiparams --optimize_out=./ch_PP-OCRv2_rec_slim_opt --valid_targets=arm --optimize_out_type=naive_buffer
# 转换方向分类器模型
./opt --model_file=./ch_ppocr_mobile_v2.0_cls_slim_infer/inference.pdmodel --param_file=./ch_ppocr_mobile_v2.0_cls_slim_infer/inference.pdiparams --optimize_out=./ch_ppocr_mobile_v2.0_cls_slim_opt --valid_targets=arm --optimize_out_type=naive_buffer
```
......@@ -186,15 +197,15 @@ wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/slim/ch_ppocr_mobile_v2.0_cls
```
准备测试图像，以`PaddleOCR/doc/imgs/11.jpg`为例，将测试的图像复制到`demo/cxx/ocr/debug/`文件夹下。
准备lite opt工具优化后的模型文件，比如使用`ch_ppocr_mobile_v2.0_det_slim_opt.nb，ch_ppocr_mobile_v2.0_rec_slim_opt.nb, ch_ppocr_mobile_v2.0_cls_slim_opt.nb`，模型文件放置在`demo/cxx/ocr/debug/`文件夹下。
准备lite opt工具优化后的模型文件，比如使用`ch_PP-OCRv2_det_slim_opt.ch_PP-OCRv2_rec_slim_rec.nb, ch_ppocr_mobile_v2.0_cls_slim_opt.nb`，模型文件放置在`demo/cxx/ocr/debug/`文件夹下。
执行完成后，ocr文件夹下将有如下文件格式：
```
demo/cxx/ocr/
|-- debug/
| |--ch_ppocr_mobile_v2.0_det_slim_opt.nb 优化后的检测模型文件
| |--ch_ppocr_mobile_v2.0_rec_slim_opt.nb 优化后的识别模型文件
| |--ch_PP-OCRv2_det_slim_opt.nb 优化后的检测模型文件
| |--ch_PP-OCRv2_rec_slim_opt.nb 优化后的识别模型文件
| |--ch_ppocr_mobile_v2.0_cls_slim_opt.nb 优化后的文字方向分类器模型文件
| |--11.jpg 待测试图像
| |--ppocr_keys_v1.txt 中文字典文件
......@@ -250,7 +261,7 @@ use_direction_classify 0 # 是否使用方向分类器，0表示不使用，1
export LD_LIBRARY_PATH=${PWD}:$LD_LIBRARY_PATH
# 开始使用，ocr_db_crnn可执行文件的使用方式为:
# ./ocr_db_crnn 检测模型文件 方向分类器模型文件 识别模型文件 测试图像路径 字典文件路径
./ocr_db_crnn ch_ppocr_mobile_v2.0_det_slim_opt.nb ch_ppocr_mobile_v2.0_rec_slim_opt.nb ch_ppocr_mobile_v2.0_cls_slim_opt.nb ./11.jpg ppocr_keys_v1.txt
./ocr_db_crnn ch_PP-OCRv2_det_slim_opt.nb ch_PP-OCRv2_rec_slim_opt.nb ch_ppocr_mobile_v2.0_cls_slim_opt.nb ./11.jpg ppocr_keys_v1.txt
```
如果对代码做了修改，则需要重新编译并push到手机上。
......
deploy/lite/readme_en.md
View file @ 3df64d50
- [Tutorial of PaddleOCR Mobile deployment](#tutorial-of-paddleocr-mobile-deployment)
- [1. Preparation](#1-preparation)
- [Preparation environment](#preparation-environment)
- [1.1 Prepare the cross-compilation environment](#11-prepare-the-cross-compilation-environment)
- [1.2 Prepare Paddle-Lite library](#12-prepare-paddle-lite-library)
- [2 Run](#2-run)
- [2.1 Inference Model Optimization](#21-inference-model-optimization)
- [2.2 Run optimized model on Phone](#22-run-optimized-model-on-phone)
- [注意：](#注意)
- [FAQ](#faq)
# Tutorial of PaddleOCR Mobile deployment
This tutorial will introduce how to use [Paddle Lite](https://github.com/PaddlePaddle/Paddle-Lite) to deploy PaddleOCR ultra-lightweight Chinese and English detection models on mobile phones.
......@@ -28,17 +39,17 @@ There are two ways to obtain the Paddle-Lite library：
| Platform | Paddle-Lite library download link |
|---|---|
|Android|[arm7](https://github.com/PaddlePaddle/Paddle-Lite/releases/download/v2.9/inference_lite_lib.android.armv7.gcc.c++_shared.with_extra.with_cv.tar.gz) / [arm8](https://github.com/PaddlePaddle/Paddle-Lite/releases/download/v2.9/inference_lite_lib.android.armv8.gcc.c++_shared.with_extra.with_cv.tar.gz)|
|IOS|[arm7](https://github.com/PaddlePaddle/Paddle-Lite/releases/download/v2.9/inference_lite_lib.ios.armv7.with_cv.with_extra.with_log.tiny_publish.tar.gz) / [arm8](https://github.com/PaddlePaddle/Paddle-Lite/releases/download/v2.9/inference_lite_lib.ios.armv8.with_cv.with_extra.with_log.tiny_publish.tar.gz)|
|Android|[arm7](https://github.com/PaddlePaddle/Paddle-Lite/releases/download/v2.10/inference_lite_lib.android.armv7.gcc.c++_shared.with_extra.with_cv.tar.gz) / [arm8](https://github.com/PaddlePaddle/Paddle-Lite/releases/download/v2.10/inference_lite_lib.android.armv8.gcc.c++_shared.with_extra.with_cv.tar.gz)|
|IOS|[arm7](https://github.com/PaddlePaddle/Paddle-Lite/releases/download/v2.10/inference_lite_lib.ios.armv7.with_cv.with_extra.with_log.tiny_publish.tar.gz) / [arm8](https://github.com/PaddlePaddle/Paddle-Lite/releases/download/v2.10/inference_lite_lib.ios.armv8.with_cv.with_extra.with_log.tiny_publish.tar.gz)|
Note: 1. The above Paddle-Lite library is compiled from the Paddle-Lite 2.9 branch. For more information about Paddle-Lite 2.9, please refer to [link](https://github.com/PaddlePaddle/Paddle-Lite/releases/tag/v2.9).
Note: 1. The above Paddle-Lite library is compiled from the Paddle-Lite 2.10 branch. For more information about Paddle-Lite 2.10, please refer to [link](https://github.com/PaddlePaddle/Paddle-Lite/releases/tag/v2.10).
- 2. [Recommended] Compile Paddle-Lite to get the prediction library. The compilation method of Paddle-Lite is as follows：
```
git clone https://github.com/PaddlePaddle/Paddle-Lite.git
cd Paddle-Lite
# Switch to Paddle-Lite release/v2.8 stable branch
git checkout release/v2.8
# Switch to Paddle-Lite release/v2.10 stable branch
git checkout release/v2.10
./lite/tools/build_android.sh --arch=armv8 --with_cv=ON --with_extra=ON
```
......@@ -87,10 +98,10 @@ The following table also provides a series of models that can be deployed on mob
|Version|Introduction|Model size|Detection model|Text Direction model|Recognition model|Paddle-Lite branch|
|---|---|---|---|---|---|---|
|V2.0|extra-lightweight chinese OCR optimized model|7.8M|[download link](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_det_opt.nb)|[download link](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_cls_opt.nb)|[download link](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_rec_opt.nb)|v2.9|
|V2.0(slim)|extra-lightweight chinese OCR optimized model|3.3M|[download link](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_det_slim_opt.nb)|[download link](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_cls_slim_opt.nb)|[download link](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_rec_slim_opt.nb)|v2.9|
|PP-OCRv2|extra-lightweight chinese OCR optimized model|11M|[download link](https://paddleocr.bj.bcebos.com/PP-OCRv2/lite/ch_PP-OCRv2_det_infer_opt.nb)|[download link](https://paddleocr.bj.bcebos.com/PP-OCRv2/lite/ch_ppocr_mobile_v2.0_cls_infer_opt.nb)|[download link](https://paddleocr.bj.bcebos.com/PP-OCRv2/lite/ch_PP-OCRv2_rec_infer_opt.nb)|v2.10|
|PP-OCRv2(slim)|extra-lightweight chinese OCR optimized model|4.6M|[download link](https://paddleocr.bj.bcebos.com/PP-OCRv2/lite/ch_PP-OCRv2_det_slim_opt.nb)|[download link](https://paddleocr.bj.bcebos.com/PP-OCRv2/lite/ch_ppocr_mobile_v2.0_cls_slim_opt.nb)|[download link](https://paddleocr.bj.bcebos.com/PP-OCRv2/lite/ch_PP-OCRv2_rec_slim_opt.nb)|v2.10|
If you directly use the model in the above table for deployment, you can skip the following steps and directly read [Section 2.2](#2.2 Run optimized model on Phone).
If you directly use the model in the above table for deployment, you can skip the following steps and directly read [Section 2.2](#2.2-Run-optimized-model-on-Phone).
If the model to be deployed is not in the above table, you need to follow the steps below to obtain the optimized model.
......@@ -98,7 +109,7 @@ The `opt` tool can be obtained by compiling Paddle Lite.
```
git clone https://github.com/PaddlePaddle/Paddle-Lite.git
cd Paddle-Lite
git checkout release/v2.9
git checkout release/v2.10
./lite/tools/build.sh build_optimize_tool
```
......@@ -124,22 +135,22 @@ cd build.opt/lite/api/
The following takes the ultra-lightweight Chinese model of PaddleOCR as an example to introduce the use of the compiled opt file to complete the conversion of the inference model to the Paddle-Lite optimized model
```
# [Recommendation] Download the Chinese and English inference model of PaddleOCR V2.0
wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/slim/ch_ppocr_mobile_v2.0_det_slim_infer.tar && tar xf ch_ppocr_mobile_v2.0_det_slim_infer.tar
wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/slim/ch_ppocr_mobile_v2.0_rec_slim_infer.tar && tar xf ch_ppocr_mobile_v2.0_rec_slim_infer.tar
# 【[Recommendation] Download the Chinese and English inference model of PP-OCRv2
wget https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_det_slim_quant_infer.tar && tar xf ch_PP-OCRv2_det_slim_quant_infer.tar
wget https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_slim_quant_infer.tar && tar xf ch_PP-OCRv2_rec_slim_quant_infer.tar
wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/slim/ch_ppocr_mobile_v2.0_cls_slim_infer.tar && tar xf ch_ppocr_mobile_v2.0_cls_slim_infer.tar
# Convert V2.0 detection model
./opt --model_file=./ch_ppocr_mobile_v2.0_det_slim_infer/inference.pdmodel --param_file=./ch_ppocr_mobile_v2.0_det_slim_infer/inference.pdiparams --optimize_out=./ch_ppocr_mobile_v2.0_det_slim_opt --valid_targets=arm --optimize_out_type=naive_buffer
# Convert V2.0 recognition model
./opt --model_file=./ch_ppocr_mobile_v2.0_rec_slim_infer/inference.pdmodel --param_file=./ch_ppocr_mobile_v2.0_rec_slim_infer/inference.pdiparams --optimize_out=./ch_ppocr_mobile_v2.0_rec_slim_opt --valid_targets=arm --optimize_out_type=naive_buffer
# Convert V2.0 angle classifier model
# Convert detection model
./opt --model_file=./ch_PP-OCRv2_det_slim_quant_infer/inference.pdmodel --param_file=./ch_PP-OCRv2_det_slim_quant_infer/inference.pdiparams --optimize_out=./ch_PP-OCRv2_det_slim_opt --valid_targets=arm --optimize_out_type=naive_buffer
# Convert recognition model
./opt --model_file=./ch_PP-OCRv2_rec_slim_quant_infer/inference.pdmodel --param_file=./ch_PP-OCRv2_rec_slim_quant_infer/inference.pdiparams --optimize_out=./ch_PP-OCRv2_rec_slim_opt --valid_targets=arm --optimize_out_type=naive_buffer
# Convert angle classifier model
./opt --model_file=./ch_ppocr_mobile_v2.0_cls_slim_infer/inference.pdmodel --param_file=./ch_ppocr_mobile_v2.0_cls_slim_infer/inference.pdiparams --optimize_out=./ch_ppocr_mobile_v2.0_cls_slim_opt --valid_targets=arm --optimize_out_type=naive_buffer
```
After the conversion is successful, there will be more files ending with `.nb` in the inference model directory, which is the successfully converted model file.
<a name="2.2 Run optimized model on Phone"></a>
<a name="2.2-Run-optimized-model-on-Phone"></a>
### 2.2 Run optimized model on Phone
Some preparatory work is required first.
......@@ -194,8 +205,8 @@ The structure of the OCR demo is as follows after the above command is executed:
```
demo/cxx/ocr/
|-- debug/
| |--ch_ppocr_mobile_v2.0_det_slim_opt.nb Detection model
| |--ch_ppocr_mobile_v2.0_rec_slim_opt.nb Recognition model
| |--ch_PP-OCRv2_det_slim_opt.nb Detection model
| |--ch_PP-OCRv2_rec_slim_opt.nb Recognition model
| |--ch_ppocr_mobile_v2.0_cls_slim_opt.nb Text direction classification model
| |--11.jpg Image for OCR
| |--ppocr_keys_v1.txt Dictionary file
......@@ -249,7 +260,7 @@ After the above steps are completed, you can use adb to push the file to the pho
export LD_LIBRARY_PATH=${PWD}:$LD_LIBRARY_PATH
# The use of ocr_db_crnn is:
# ./ocr_db_crnn Detection model file Orientation classifier model file Recognition model file Test image path Dictionary file path
./ocr_db_crnn ch_ppocr_mobile_v2.0_det_opt.nb ch_ppocr_mobile_v2.0_rec_opt.nb ch_ppocr_mobile_v2.0_cls_opt.nb ./11.jpg ppocr_keys_v1.txt
./ocr_db_crnn ch_PP-OCRv2_det_slim_opt.nb ch_PP-OCRv2_rec_slim_opt.nb ch_ppocr_mobile_v2.0_cls_opt.nb ./11.jpg ppocr_keys_v1.txt
```
If you modify the code, you need to recompile and push to the phone.
......
doc/doc_ch/models_list.md
View file @ 3df64d50
......@@ -6,13 +6,14 @@
> 3. 本文档提供的是PPOCR自研模型列表，更多基于公开数据集的算法介绍与预训练模型可以参考：[算法概览文档](./algorithm_overview.md)。
- [1. 文本检测模型](#文本检测模型)
- [2. 文本识别模型](#文本识别模型)
- [2.1 中文识别模型](#中文识别模型)
- [2.2 英文识别模型](#英文识别模型)
- [2.3 多语言识别模型](#多语言识别模型)
- [3. 文本方向分类模型](#文本方向分类模型)
- [4. Paddle-Lite 模型](#Paddle-Lite模型)
- [PP-OCR系列模型列表（V2.1，2021年9月6日更新）](#pp-ocr系列模型列表v212021年9月6日更新)
- [1. 文本检测模型](#1-文本检测模型)
- [2. 文本识别模型](#2-文本识别模型)
- [2.1 中文识别模型](#21-中文识别模型)
- [2.2 英文识别模型](#22-英文识别模型)
- [2.3 多语言识别模型（更多语言持续更新中...）](#23-多语言识别模型更多语言持续更新中)
- [3. 文本方向分类模型](#3-文本方向分类模型)
- [4. Paddle-Lite 模型](#4-paddle-lite-模型)
PaddleOCR提供的可下载模型包括`推理模型`、`训练模型`、`预训练模型`、`slim模型`，模型区别说明如下：
......@@ -100,6 +101,8 @@ PaddleOCR提供的可下载模型包括`推理模型`、`训练模型`、`预训
|模型版本|模型简介|模型大小|检测模型|文本方向分类模型|识别模型|Paddle-Lite版本|
|---|---|---|---|---|---|---|
|PP-OCRv2|蒸馏版超轻量中文OCR移动端模型|11M|[下载地址](https://paddleocr.bj.bcebos.com/PP-OCRv2/lite/ch_PP-OCRv2_det_infer_opt.nb)|[下载地址](https://paddleocr.bj.bcebos.com/PP-OCRv2/lite/ch_ppocr_mobile_v2.0_cls_infer_opt.nb)|[下载地址](https://paddleocr.bj.bcebos.com/PP-OCRv2/lite/ch_PP-OCRv2_rec_infer_opt.nb)|v2.10|
|PP-OCRv2(slim)|蒸馏版超轻量中文OCR移动端模型|4.6M|[下载地址](https://paddleocr.bj.bcebos.com/PP-OCRv2/lite/ch_PP-OCRv2_det_slim_opt.nb)|[下载地址](https://paddleocr.bj.bcebos.com/PP-OCRv2/lite/ch_ppocr_mobile_v2.0_cls_slim_opt.nb)|[下载地址](https://paddleocr.bj.bcebos.com/PP-OCRv2/lite/ch_PP-OCRv2_rec_slim_opt.nb)|v2.10|
|PP-OCRv2|蒸馏版超轻量中文OCR移动端模型|11M|[下载地址](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_det_infer_opt.nb)|[下载地址](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_cls_opt.nb)|[下载地址](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_infer_opt.nb)|v2.9|
|PP-OCRv2(slim)|蒸馏版超轻量中文OCR移动端模型|4.9M|[下载地址](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_det_slim_opt.nb)|[下载地址](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_cls_slim_opt.nb)|[下载地址](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_slim_opt.nb)|v2.9|
|V2.0|ppocr_v2.0超轻量中文OCR移动端模型|7.8M|[下载地址](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_det_opt.nb)|[下载地址](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_cls_opt.nb)|[下载地址](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_rec_opt.nb)|v2.9|
......
doc/doc_en/models_list_en.md
View file @ 3df64d50
......@@ -94,6 +94,8 @@ For more supported languages, please refer to : [Multi-language model](./multi_l
## 4. Paddle-Lite Model
|Version|Introduction|Model size|Detection model|Text Direction model|Recognition model|Paddle-Lite branch|
|---|---|---|---|---|---|---|
|PP-OCRv2|extra-lightweight chinese OCR optimized model|11M|[download link](https://paddleocr.bj.bcebos.com/PP-OCRv2/lite/ch_PP-OCRv2_det_infer_opt.nb)|[download link](https://paddleocr.bj.bcebos.com/PP-OCRv2/lite/ch_ppocr_mobile_v2.0_cls_infer_opt.nb)|[download link](https://paddleocr.bj.bcebos.com/PP-OCRv2/lite/ch_PP-OCRv2_rec_infer_opt.nb)|v2.10|
|PP-OCRv2(slim)|extra-lightweight chinese OCR optimized model|4.6M|[download link](https://paddleocr.bj.bcebos.com/PP-OCRv2/lite/ch_PP-OCRv2_det_slim_opt.nb)|[download link](https://paddleocr.bj.bcebos.com/PP-OCRv2/lite/ch_ppocr_mobile_v2.0_cls_slim_opt.nb)|[download link](https://paddleocr.bj.bcebos.com/PP-OCRv2/lite/ch_PP-OCRv2_rec_slim_opt.nb)|v2.10|
|PP-OCRv2|extra-lightweight chinese OCR optimized model|11M|[download link](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_det_infer_opt.nb)|[download link](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_cls_opt.nb)|[download link](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_infer_opt.nb)|v2.9|
|PP-OCRv2(slim)|extra-lightweight chinese OCR optimized model|4.9M|[download link](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_det_slim_opt.nb)|[download link](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_cls_slim_opt.nb)|[download link](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_slim_opt.nb)|v2.9|
|V2.0|ppocr_v2.0 extra-lightweight chinese OCR optimized model|7.8M|[download link](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_det_opt.nb)|[download link](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_cls_opt.nb)|[download link](https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_ppocr_mobile_v2.0_rec_opt.nb)|v2.9|
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment