diff --git a/model_zoo/official/lite/image_classification/README.en.md b/model_zoo/official/lite/image_classification/README.en.md index fd3c4f59d3..7c5d94d4fd 100644 --- a/model_zoo/official/lite/image_classification/README.en.md +++ b/model_zoo/official/lite/image_classification/README.en.md @@ -1,52 +1,67 @@ -## Demo of Image Classification +# Demo of Image Classification The following describes how to use the MindSpore Lite C++ APIs (Android JNIs) and MindSpore Lite image classification models to perform on-device inference, classify the content captured by a device camera, and display the most possible classification result on the application's image preview screen. -### Running Dependencies +## Running Dependencies - Android Studio 3.2 or later (Android 4.0 or later is recommended.) -- Native development kit (NDK) 21.3 -- CMake 3.10.2 [CMake](https://cmake.org/download) -- Android software development kit (SDK) 26 or later -- JDK 1.8 or later -### Building and Running +## Building and Running -1. Load the sample source code to Android Studio and install the corresponding SDK. (After the SDK version is specified, Android Studio automatically installs the SDK.) +1. Load the sample source code to Android Studio. ![start_home](images/home.png) - Start Android Studio, click `File > Settings > System Settings > Android SDK`, and select the corresponding SDK. As shown in the following figure, select an SDK and click `OK`. Android Studio automatically installs the SDK. + Start Android Studio, click `File > Settings > System Settings > Android SDK`, and select the corresponding `SDK Tools`. As shown in the following figure, select an SDK and click `OK`. Android Studio automatically installs the SDK. - ![start_sdk](images/sdk_management.png) + ![start_sdk](images/sdk_management.jpg) - If you have any Android Studio configuration problem when trying this demo, please refer to item 4 to resolve it. + > Android SDK Tools is the default installation. You can see this by unchecking the `Hide Obsolete Packages`box. + > + > If you have any Android Studio configuration problem when trying this demo, please refer to item 4 to resolve it. -2. Connect to an Android device and runs the image classification application. +2. Connect to an Android device and runs this application. Connect to the Android device through a USB cable for debugging. Click `Run 'app'` to run the sample project on your device. ![run_app](images/run_app.PNG) - For details about how to connect the Android Studio to a device for debugging, see . - - The mobile phone needs to be turn on "USB debugging mode" before Android Studio can recognize the mobile phone. Huawei mobile phones generally turn on "USB debugging model" in Settings > system and update > developer Options > USB debugging. + > Android Studio will automatically download MindSpore Lite, model files and other dependencies during the compilation process. Please be patient during this process. + > + > For details about how to connect the Android Studio to a device for debugging, see . + > + > The mobile phone needs to be turn on "USB debugging mode" before Android Studio can recognize the mobile phone. Huawei mobile phones generally turn on "USB debugging model" in Settings -> system and update -> developer Options -> USB debugging. 3. Continue the installation on the Android device. After the installation is complete, you can view the content captured by a camera and the inference result. ![result](images/app_result.jpg) -4. The solutions of Android Studio configuration problems: +4. The solutions of configuration problems: + + 4.1 Problems of NDK, CMake, JDK Tools: + + If the tools installed in Android Studio are not recognized, you can re-download and install them from the corresponding official website, and configure the path. - | | Warning | Solution | - | ---- | ------------------------------------------------------------ | ------------------------------------------------------------ | - | 1 | Gradle sync failed: NDK not configured. | Specify the installed ndk directory in local.properties：ndk.dir={ndk的安装目录} | - | 2 | Requested NDK version did not match the version requested by ndk.dir | Manually download corresponding [NDK Version](https://developer.android.com/ndk/downloads)，and specify the sdk directory in Project Structure - Android NDK location.（You can refer to the figure below.） | - | 3 | This version of Android Studio cannot open this project, please retry with Android Studio or newer. | Update Android Studio Version in Tools - help - Checkout for Updates. | - | 4 | SSL peer shut down incorrectly | Run this demo again. | + - NDK >= 21.3 [NDK](https://developer.android.google.cn/ndk/downloads?hl=zh-cn) + - CMake >= 3.10.2 [CMake](https://cmake.org/download) + - Android SDK >= 26 [SDK](https://developer.microsoft.com/zh-cn/windows/downloads/windows-10-sdk/) + - JDK >= 1.8 [JDK](https://www.oracle.com/cn/java/technologies/javase/javase-jdk8-downloads.html) ![project_structure](images/project_structure.png) + 4.2 NDK version does not match: + + Open `Android SDK`, click `Show Package Details`, and select the appropriate NDK version according to the error message. + ![NDK_version](images/NDK_version.jpg) + + 4.3 Problem of Android Studio version: + + Update the Android Studio version in `Toolbar - Help - Checkout for Updates`. + + 4.4 Gradle dependencies installed too slowly: + + As shown in the picture, open the Demo root directory `build. Gradle` file, then add huawei mirror source address: `maven {url 'https://developer.huawei.com/repo/'}`, modify the classpath to 4.0.0 and click ` sync ` . Once the download is complete, restore the classpath version and synchronize it again. + ## Detailed Description of the Sample Program This image classification sample program on the Android device includes a Java layer and a JNI layer. At the Java layer, the Android Camera 2 API is used to enable a camera to obtain image frames and process images. At the JNI layer, the model inference process is completed in [Runtime](https://www.mindspore.cn/tutorial/lite/en/master/use/runtime.html). diff --git a/model_zoo/official/lite/image_classification/README.md b/model_zoo/official/lite/image_classification/README.md index a0bf8f8308..270b51127c 100644 --- a/model_zoo/official/lite/image_classification/README.md +++ b/model_zoo/official/lite/image_classification/README.md @@ -1,39 +1,37 @@ -## MindSpore Lite 端侧图像分类demo（Android） +# MindSpore Lite 端侧图像分类demo（Android） 本示例程序演示了如何在端侧利用MindSpore Lite C++ API（Android JNI）以及MindSpore Lite 图像分类模型完成端侧推理，实现对设备摄像头捕获的内容进行分类，并在App图像预览界面中显示出最可能的分类结果。 -### 运行依赖 +## 运行依赖 - Android Studio >= 3.2 (推荐4.0以上版本) -- NDK 21.3 -- CMake 3.10.2 [CMake](https://cmake.org/download) -- Android SDK >= 26 -- JDK >= 1.8 -### 构建与运行 +## 构建与运行 -1. 在Android Studio中加载本示例源码，并安装相应的SDK（指定SDK版本后，由Android Studio自动安装）。 +1. 在Android Studio中加载本示例源码。 ![start_home](images/home.png) - 启动Android Studio后，点击`File->Settings->System Settings->Android SDK`，勾选相应的SDK。如下图所示，勾选后，点击`OK`，Android Studio即可自动安装SDK。 + 启动Android Studio后，点击`File->Settings->System Settings->Android SDK`，勾选相应的`SDK Tools`。如下图所示，勾选后，点击`OK`，Android Studio即可自动安装SDK。 - ![start_sdk](images/sdk_management.png) + ![start_sdk](images/sdk_management.jpg) - 使用过程中若出现Android Studio配置问题，可参考第4项解决。 + > Android SDK Tools为默认安装项，取消`Hide Obsolete Packages`选框之后可看到。 + > + > 使用过程中若出现问题，可参考第4项解决。 -2. 连接Android设备，运行图像分类应用程序。 +2. 连接Android设备，运行该应用程序。 - 通过USB连接Android设备调试，点击`Run 'app'`即可在您的设备上运行本示例项目。 + 通过USB连接Android手机。待成功识别到设备后，点击`Run 'app'`即可在您的手机上运行本示例项目。 > 编译过程中Android Studio会自动下载MindSpore Lite、模型文件等相关依赖项，编译过程需做耐心等待。 + > + > Android Studio连接设备调试操作，可参考。 + > + > 手机需开启“USB调试模式”，Android Studio 才能识别到手机。 华为手机一般在设置->系统和更新->开发人员选项->USB调试中开始“USB调试模型”。 ![run_app](images/run_app.PNG) - Android Studio连接设备调试操作，可参考。 - - 手机需开启“USB调试模式”，Android Studio 才能识别到手机。 华为手机一般在设置->系统和更新->开发人员选项->USB调试中开始“USB调试模型”。 - 3. 在Android设备上，点击“继续安装”，安装完即可查看到设备摄像头捕获的内容和推理结果。 ![install](images/install.jpg) @@ -42,16 +40,32 @@ ![result](images/app_result.jpg) -4. Android Studio 配置问题解决方案可参考下表： +4. Demo部署问题解决方案。 + + 4.1 NDK、CMake、JDK等工具问题： + + 如果Android Studio内安装的工具出现无法识别等问题，可重新从相应官网下载和安装，并配置路径。 + + - NDK >= 21.3 [NDK](https://developer.android.google.cn/ndk/downloads?hl=zh-cn) + - CMake >= 3.10.2 [CMake](https://cmake.org/download) + - Android SDK >= 26 [SDK](https://developer.microsoft.com/zh-cn/windows/downloads/windows-10-sdk/) + - JDK >= 1.8 [JDK](https://www.oracle.com/cn/java/technologies/javase/javase-jdk8-downloads.html) + + ![project_structure](images/project_structure.png) + + 4.2 NDK版本不匹配问题： + + 打开`Android SDK`，点击`Show Package Details`，根据报错信息选择安装合适的NDK版本。 + ![NDK_version](images/NDK_version.jpg) + + 4.3 Android Studio版本问题： + + 在`工具栏-help-Checkout for Updates`中更新Android Studio版本。 - | | 报错 | 解决方案 | - | ---- | ------------------------------------------------------------ | ------------------------------------------------------------ | - | 1 | Gradle sync failed: NDK not configured. | 在local.properties中指定安装的ndk目录：ndk.dir={ndk的安装目录} | - | 2 | Requested NDK version did not match the version requested by ndk.dir | 可手动下载相应的[NDK版本](https://developer.android.com/ndk/downloads?hl=zh-cn)，并在Project Structure - Android NDK location设置中指定SDK的位置（可参考下图完成） | - | 3 | This version of Android Studio cannot open this project, please retry with Android Studio or newer. | 在工具栏-help-Checkout for Updates中更新版本 | - | 4 | SSL peer shut down incorrectly | 重新构建 | + 4.4 Gradle下依赖项安装过慢问题： - ![project_structure](images/project_structure.png) + 如图所示， 打开Demo根目录下`build.gradle`文件，加入华为镜像源地址：`maven {url 'https://developer.huawei.com/repo/'}`，修改classpath为4.0.0，点击`sync`进行同步。下载完成后，将classpath版本复原，再次进行同步。 + ![maven](images/maven.jpg) ## 示例程序详细说明 diff --git a/model_zoo/official/lite/image_classification/images/NDK_version.jpg b/model_zoo/official/lite/image_classification/images/NDK_version.jpg new file mode 100644 index 0000000000..50af23b0db Binary files /dev/null and b/model_zoo/official/lite/image_classification/images/NDK_version.jpg differ diff --git a/model_zoo/official/lite/image_classification/images/maven.jpg b/model_zoo/official/lite/image_classification/images/maven.jpg new file mode 100644 index 0000000000..eee21bb1df Binary files /dev/null and b/model_zoo/official/lite/image_classification/images/maven.jpg differ diff --git a/model_zoo/official/lite/image_classification/images/sdk_management.jpg b/model_zoo/official/lite/image_classification/images/sdk_management.jpg new file mode 100644 index 0000000000..4051b9bab3 Binary files /dev/null and b/model_zoo/official/lite/image_classification/images/sdk_management.jpg differ diff --git a/model_zoo/official/lite/image_classification/images/sdk_management.png b/model_zoo/official/lite/image_classification/images/sdk_management.png deleted file mode 100644 index faf694bd2e..0000000000 Binary files a/model_zoo/official/lite/image_classification/images/sdk_management.png and /dev/null differ diff --git a/model_zoo/official/lite/image_segmentation/README.en.md b/model_zoo/official/lite/image_segmentation/README.en.md index 7b5098b1ab..18ff452fd9 100644 --- a/model_zoo/official/lite/image_segmentation/README.en.md +++ b/model_zoo/official/lite/image_segmentation/README.en.md @@ -1,50 +1,69 @@ -## Demo of Image Segmentation +# Demo of Image Segmentation The following describes how to use the MindSpore Lite JAVA APIs and MindSpore Lite image segmentation models to perform on-device inference, classify the content captured by a device camera, and display the most possible segmentation result on the application's image preview screen. -### Running Dependencies +## Running Dependencies - Android Studio 3.2 or later (Android 4.0 or later is recommended.) -- Native development kit (NDK) 21.3 -- CMake 3.10.2 [CMake](https://cmake.org/download) -- Android software development kit (SDK) 26 or later -- JDK 1.8 or later -### Building and Running +## Building and Running -1. Load the sample source code to Android Studio and install the corresponding SDK. (After the SDK version is specified, Android Studio automatically installs the SDK.) +1. Load the sample source code to Android Studio. ![start_home](images/home.png) - Start Android Studio, click `File > Settings > System Settings > Android SDK`, and select the corresponding SDK. As shown in the following figure, select an SDK and click `OK`. Android Studio automatically installs the SDK. + Start Android Studio, click `File > Settings > System Settings > Android SDK`, and select the corresponding `SDK Tools`. As shown in the following figure, select an SDK and click `OK`. Android Studio automatically installs the SDK. - ![start_sdk](images/sdk_management.png) + ![start_sdk](images/sdk_management.jpg) - If you have any Android Studio configuration problem when trying this demo, please refer to item 5 to resolve it. + > Android SDK Tools is the default installation. You can see this by unchecking the `Hide Obsolete Packages`box. + > + > If you have any Android Studio configuration problem when trying this demo, please refer to item 4 to resolve it. -2. Connect to an Android device and runs the image segmentation application. +2. Connect to an Android device and runs this application. Connect to the Android device through a USB cable for debugging. Click `Run 'app'` to run the sample project on your device. ![run_app](images/run_app.PNG) - For details about how to connect the Android Studio to a device for debugging, see . - - The mobile phone needs to be turn on "USB debugging mode" before Android Studio can recognize the mobile phone. Huawei mobile phones generally turn on "USB debugging model" in Settings > system and update > developer Options > USB debugging. + > Android Studio will automatically download MindSpore Lite, model files and other dependencies during the compilation process. Please be patient during this process. + > + > For details about how to connect the Android Studio to a device for debugging, see . + > + > The mobile phone needs to be turn on "USB debugging mode" before Android Studio can recognize the mobile phone. Huawei mobile phones generally turn on "USB debugging model" in Settings -> system and update -> developer Options -> USB debugging. 3. Continue the installation on the Android device. After the installation is complete, you can view the content captured by a camera and the inference result. -4. The solutions of Android Studio configuration problems: + ![result](images/app_result.jpg) + +4. The solutions of configuration problems: + + 4.1 Problems of NDK, CMake, JDK Tools: - | | Warning | Solution | - | ---- | ------------------------------------------------------------ | ------------------------------------------------------------ | - | 1 | Gradle sync failed: NDK not configured. | Specify the installed ndk directory in local.properties：ndk.dir={ndk的安装目录} | - | 2 | Requested NDK version did not match the version requested by ndk.dir | Manually download corresponding [NDK Version](https://developer.android.com/ndk/downloads)，and specify the sdk directory in Project Structure - Android NDK location.（You can refer to the figure below.） | - | 3 | This version of Android Studio cannot open this project, please retry with Android Studio or newer. | Update Android Studio Version in Tools - help - Checkout for Updates. | - | 4 | SSL peer shut down incorrectly | Run this demo again. | + If the tools installed in Android Studio are not recognized, you can re-download and install them from the corresponding official website, and configure the path. + + - NDK >= 21.3 [NDK](https://developer.android.google.cn/ndk/downloads?hl=zh-cn) + - CMake >= 3.10.2 [CMake](https://cmake.org/download) + - Android SDK >= 26 [SDK](https://developer.microsoft.com/zh-cn/windows/downloads/windows-10-sdk/) + - JDK >= 1.8 [JDK](https://www.oracle.com/cn/java/technologies/javase/javase-jdk8-downloads.html) ![project_structure](images/project_structure.png) + 4.2 NDK version does not match: + + Open `Android SDK`, click `Show Package Details`, and select the appropriate NDK version according to the error message. + ![NDK_version](images/NDK_version.jpg) + + 4.3 Problem of Android Studio version: + + Update the Android Studio version in `Toolbar - Help - Checkout for Updates`. + + 4.4 Gradle dependencies installed too slowly: + + As shown in the picture, open the Demo root directory `build. Gradle` file, then add huawei mirror source address: `maven {url 'https://developer.huawei.com/repo/'}`, modify the classpath to 4.0.0 and click ` sync ` . Once the download is complete, restore the classpath version and synchronize it again. + + ![maven](images/maven.jpg) + ## Detailed Description of the Sample Program This image segmentation sample program on the Android device is implemented through Java. At the Java layer, the Android Camera 2 API is used to enable a camera to obtain image frames and process images. Then Java API is called to infer.[Runtime](https://www.mindspore.cn/tutorial/lite/en/master/use/runtime.html). diff --git a/model_zoo/official/lite/image_segmentation/README.md b/model_zoo/official/lite/image_segmentation/README.md index d2583649e9..53b118cee6 100644 --- a/model_zoo/official/lite/image_segmentation/README.md +++ b/model_zoo/official/lite/image_segmentation/README.md @@ -1,53 +1,71 @@ -## MindSpore Lite 端侧图像分割demo（Android） +# MindSpore Lite 端侧图像分割demo（Android） 本示例程序演示了如何在端侧利用MindSpore Lite Java API 以及MindSpore Lite 图像分割模型完成端侧推理，实现对设备摄像头捕获的内容进行分割，并在App图像预览界面中显示出最可能的分割结果。 -### 运行依赖 +## 运行依赖 - Android Studio >= 3.2 (推荐4.0以上版本) -- NDK 21.3 -- CMake 3.10.2 [CMake](https://cmake.org/download) -- Android SDK >= 26 -- JDK >= 1.8 -### 构建与运行 +## 构建与运行 -1. 在Android Studio中加载本示例源码，并安装相应的SDK（指定SDK版本后，由Android Studio自动安装）。 +1. 在Android Studio中加载本示例源码。 ![start_home](images/home.png) - 启动Android Studio后，点击`File->Settings->System Settings->Android SDK`，勾选相应的SDK。如下图所示，勾选后，点击`OK`，Android Studio即可自动安装SDK。 + 启动Android Studio后，点击`File->Settings->System Settings->Android SDK`，勾选相应的`SDK Tools`。如下图所示，勾选后，点击`OK`，Android Studio即可自动安装SDK。 - ![start_sdk](images/sdk_management.png) + ![start_sdk](images/sdk_management.jpg) - 使用过程中若出现Android Studio配置问题，可参考第4项解决。 + > Android SDK Tools为默认安装项，取消`Hide Obsolete Packages`选框之后可看到。 + > + > 使用过程中若出现问题，可参考第4项解决。 -2. 连接Android设备，运行图像分割应用程序。 +2. 连接Android设备，运行该应用程序。 - 通过USB连接Android设备调试，点击`Run 'app'`即可在您的设备上运行本示例项目。 + 通过USB连接Android手机。待成功识别到设备后，点击`Run 'app'`即可在您的手机上运行本示例项目。 > 编译过程中Android Studio会自动下载MindSpore Lite、模型文件等相关依赖项，编译过程需做耐心等待。 + > + > Android Studio连接设备调试操作，可参考。 + > + > 手机需开启“USB调试模式”，Android Studio 才能识别到手机。 华为手机一般在设置->系统和更新->开发人员选项->USB调试中开始“USB调试模型”。 ![run_app](images/run_app.PNG) - Android Studio连接设备调试操作，可参考。 - - 手机需开启“USB调试模式”，Android Studio 才能识别到手机。 华为手机一般在设置->系统和更新->开发人员选项->USB调试中开始“USB调试模型”。 - 3. 在Android设备上，点击“继续安装”，安装完即可查看到设备摄像头捕获的内容和推理结果。 ![install](images/install.jpg) -4. Android Studio 配置问题解决方案可参考下表： + 如下图所示，识别出的概率最高的物体是植物。 + + ![result](images/app_result.jpg) + +4. Demo部署问题解决方案。 + + 4.1 NDK、CMake、JDK等工具问题： + + 如果Android Studio内安装的工具出现无法识别等问题，可重新从相应官网下载和安装，并配置路径。 + + - NDK >= 21.3 [NDK](https://developer.android.google.cn/ndk/downloads?hl=zh-cn) + - CMake >= 3.10.2 [CMake](https://cmake.org/download) + - Android SDK >= 26 [SDK](https://developer.microsoft.com/zh-cn/windows/downloads/windows-10-sdk/) + - JDK >= 1.8 [JDK](https://www.oracle.com/cn/java/technologies/javase/javase-jdk8-downloads.html) + + ![project_structure](images/project_structure.png) + + 4.2 NDK版本不匹配问题： + + 打开`Android SDK`，点击`Show Package Details`，根据报错信息选择安装合适的NDK版本。 + ![NDK_version](images/NDK_version.jpg) + + 4.3 Android Studio版本问题： + + 在`工具栏-help-Checkout for Updates`中更新Android Studio版本。 - | | 报错 | 解决方案 | - | ---- | ------------------------------------------------------------ | ------------------------------------------------------------ | - | 1 | Gradle sync failed: NDK not configured. | 在local.properties中指定安装的ndk目录：ndk.dir={ndk的安装目录} | - | 2 | Requested NDK version did not match the version requested by ndk.dir | 可手动下载相应的[NDK版本](https://developer.android.com/ndk/downloads?hl=zh-cn)，并在Project Structure - Android NDK location设置中指定SDK的位置（可参考下图完成） | - | 3 | This version of Android Studio cannot open this project, please retry with Android Studio or newer. | 在工具栏-help-Checkout for Updates中更新版本 | - | 4 | SSL peer shut down incorrectly | 重新构建 | + 4.4 Gradle下依赖项安装过慢问题： - ![project_structure](images/project_structure.png) + 如图所示， 打开Demo根目录下`build.gradle`文件，加入华为镜像源地址：`maven {url 'https://developer.huawei.com/repo/'}`，修改classpath为4.0.0，点击`sync`进行同步。下载完成后，将classpath版本复原，再次进行同步。 + ![maven](images/maven.jpg) ## 示例程序详细说明 diff --git a/model_zoo/official/lite/image_segmentation/images/NDK_version.jpg b/model_zoo/official/lite/image_segmentation/images/NDK_version.jpg new file mode 100644 index 0000000000..50af23b0db Binary files /dev/null and b/model_zoo/official/lite/image_segmentation/images/NDK_version.jpg differ diff --git a/model_zoo/official/lite/image_segmentation/images/maven.jpg b/model_zoo/official/lite/image_segmentation/images/maven.jpg new file mode 100644 index 0000000000..eee21bb1df Binary files /dev/null and b/model_zoo/official/lite/image_segmentation/images/maven.jpg differ diff --git a/model_zoo/official/lite/image_segmentation/images/sdk_management.jpg b/model_zoo/official/lite/image_segmentation/images/sdk_management.jpg new file mode 100644 index 0000000000..4051b9bab3 Binary files /dev/null and b/model_zoo/official/lite/image_segmentation/images/sdk_management.jpg differ diff --git a/model_zoo/official/lite/image_segmentation/images/sdk_management.png b/model_zoo/official/lite/image_segmentation/images/sdk_management.png deleted file mode 100644 index faf694bd2e..0000000000 Binary files a/model_zoo/official/lite/image_segmentation/images/sdk_management.png and /dev/null differ diff --git a/model_zoo/official/lite/object_detection/README.en.md b/model_zoo/official/lite/object_detection/README.en.md index 79238c47ae..bc2eb4a915 100644 --- a/model_zoo/official/lite/object_detection/README.en.md +++ b/model_zoo/official/lite/object_detection/README.en.md @@ -2,49 +2,68 @@ The following describes how to use the MindSpore Lite C++ APIs (Android JNIs) and MindSpore Lite object detection models to perform on-device inference, detect the content captured by a device camera, and display the most possible detection result on the application's image preview screen. -## Deploying an Application +## Running Dependencies -The following section describes how to build and execute an on-device object detecion task on MindSpore Lite. +- Android Studio 3.2 or later (Android 4.0 or later is recommended.) -### Running Dependencies +## Building and Running -- Android Studio 3.2 or later (Android 4.0 or later is recommended.) -- Native development kit (NDK) 21.3 -- CMake 3.10.2 -- Android software development kit (SDK) 26 or later -- OpenCV 4.0.0 or later (included in the sample code) +1. Load the sample source code to Android Studio. -### Building and Running + ![start_home](images/home.png) -1. Load the sample source code to Android Studio and install the corresponding SDK. (After the SDK version is specified, Android Studio automatically installs the SDK.) + Start Android Studio, click `File > Settings > System Settings > Android SDK`, and select the corresponding `SDK Tools`. As shown in the following figure, select an SDK and click `OK`. Android Studio automatically installs the SDK. - ![start_home](images/home.png) + ![start_sdk](images/sdk_management.jpg) - If you have any Android Studio configuration problem when trying this demo, please refer to item 4 to resolve it. + > Android SDK Tools is the default installation. You can see this by unchecking the `Hide Obsolete Packages`box. + > + > If you have any Android Studio configuration problem when trying this demo, please refer to item 4 to resolve it. -2. Connect to an Android device and runs the object detection application. +2. Connect to an Android device and runs this application. - Connect to the Android device through a USB cable for debugging. Click `Run 'app'` to run the sample project on your device. + Connect to the Android device through a USB cable for debugging. Click `Run 'app'` to run the sample project on your device. - ![run_app](images/project_structure.png) + ![run_app](images/run_app.PNG) - For details about how to connect the Android Studio to a device for debugging, see . + > Android Studio will automatically download MindSpore Lite, model files and other dependencies during the compilation process. Please be patient during this process. + > + > For details about how to connect the Android Studio to a device for debugging, see . + > + > The mobile phone needs to be turn on "USB debugging mode" before Android Studio can recognize the mobile phone. Huawei mobile phones generally turn on "USB debugging model" in Settings -> system and update -> developer Options -> USB debugging. 3. Continue the installation on the Android device. After the installation is complete, you can view the content captured by a camera and the inference result. - ![result](images/object_detection.png) + ![result](images/app_result.jpg) + +4. The solutions of configuration problems: -4. The solutions of Android Studio configuration problems: + 4.1 Problems of NDK, CMake, JDK Tools: - | | Warning | Solution | - | ---- | ------------------------------------------------------------ | ------------------------------------------------------------ | - | 1 | Gradle sync failed: NDK not configured. | Specify the installed ndk directory in local.properties：ndk.dir={ndk的安装目录} | - | 2 | Requested NDK version did not match the version requested by ndk.dir | Manually download corresponding [NDK Version](https://developer.android.com/ndk/downloads)，and specify the sdk directory in Project Structure - Android NDK location.（You can refer to the figure below.） | - | 3 | This version of Android Studio cannot open this project, please retry with Android Studio or newer. | Update Android Studio Version in Tools - help - Checkout for Updates. | - | 4 | SSL peer shut down incorrectly | Run this demo again. | + If the tools installed in Android Studio are not recognized, you can re-download and install them from the corresponding official website, and configure the path. + + - NDK >= 21.3 [NDK](https://developer.android.google.cn/ndk/downloads?hl=zh-cn) + - CMake >= 3.10.2 [CMake](https://cmake.org/download) + - Android SDK >= 26 [SDK](https://developer.microsoft.com/zh-cn/windows/downloads/windows-10-sdk/) + - JDK >= 1.8 [JDK](https://www.oracle.com/cn/java/technologies/javase/javase-jdk8-downloads.html) ![project_structure](images/project_structure.png) + 4.2 NDK version does not match: + + Open `Android SDK`, click `Show Package Details`, and select the appropriate NDK version according to the error message. + ![NDK_version](images/NDK_version.jpg) + + 4.3 Problem of Android Studio version: + + Update the Android Studio version in `Toolbar - Help - Checkout for Updates`. + + 4.4 Gradle dependencies installed too slowly: + + As shown in the picture, open the Demo root directory `build. Gradle` file, then add huawei mirror source address: `maven {url 'https://developer.huawei.com/repo/'}`, modify the classpath to 4.0.0 and click ` sync ` . Once the download is complete, restore the classpath version and synchronize it again. + + ![maven](images/maven.jpg) + ## Detailed Description of the Sample Program This object detection sample program on the Android device includes a Java layer and a JNI layer. At the Java layer, the Android Camera 2 API is used to enable a camera to obtain image frames and process images. At the JNI layer, the model inference process is completed . diff --git a/model_zoo/official/lite/object_detection/README.md b/model_zoo/official/lite/object_detection/README.md index 4dff47ff4a..19a9052a17 100644 --- a/model_zoo/official/lite/object_detection/README.md +++ b/model_zoo/official/lite/object_detection/README.md @@ -5,49 +5,67 @@ ## 运行依赖 - Android Studio >= 3.2 (推荐4.0以上版本) -- NDK 21.3 -- CMake 3.10 -- Android SDK >= 26 ## 构建与运行 -1. 在Android Studio中加载本示例源码，并安装相应的SDK（指定SDK版本后，由Android Studio自动安装）。 +1. 在Android Studio中加载本示例源码。 ![start_home](images/home.png) - 启动Android Studio后，点击`File->Settings->System Settings->Android SDK`，勾选相应的SDK。如下图所示，勾选后，点击`OK`，Android Studio即可自动安装SDK。 + 启动Android Studio后，点击`File->Settings->System Settings->Android SDK`，勾选相应的`SDK Tools`。如下图所示，勾选后，点击`OK`，Android Studio即可自动安装SDK。 - ![start_sdk](images/sdk_management.png) + ![start_sdk](images/sdk_management.jpg) - 使用过程中若出现Android Studio配置问题，可参考第4项解决。 + > Android SDK Tools为默认安装项，取消`Hide Obsolete Packages`选框之后可看到。 + > + > 使用过程中若出现问题，可参考第4项解决。 -2. 连接Android设备，运行目标检测示例应用程序。 +2. 连接Android设备，运行该应用程序。 - 通过USB连接Android设备调试，点击`Run 'app'`即可在你的设备上运行本示例项目。 - > 编译过程中Android Studio会自动下载MindSpore Lite、模型文件等相关依赖项，编译过程需做耐心等待。 + 通过USB连接Android手机。待成功识别到设备后，点击`Run 'app'`即可在您的手机上运行本示例项目。 - ![run_app](images/run_app.PNG) + > 编译过程中Android Studio会自动下载MindSpore Lite、模型文件等相关依赖项，编译过程需做耐心等待。 + > + > Android Studio连接设备调试操作，可参考。 + > + > 手机需开启“USB调试模式”，Android Studio 才能识别到手机。 华为手机一般在设置->系统和更新->开发人员选项->USB调试中开始“USB调试模型”。 - Android Studio连接设备调试操作，可参考。 + ![run_app](images/run_app.PNG) 3. 在Android设备上，点击“继续安装”，安装完即可查看到设备摄像头捕获的内容和推理结果。 ![install](images/install.jpg) - 如下图所示，检测出图中内容是鼠标。 + 如下图所示，识别出的概率最高的物体是植物。 + + ![result](images/app_result.jpg) + +4. Demo部署问题解决方案。 + + 4.1 NDK、CMake、JDK等工具问题： + + 如果Android Studio内安装的工具出现无法识别等问题，可重新从相应官网下载和安装，并配置路径。 + + - NDK >= 21.3 [NDK](https://developer.android.google.cn/ndk/downloads?hl=zh-cn) + - CMake >= 3.10.2 [CMake](https://cmake.org/download) + - Android SDK >= 26 [SDK](https://developer.microsoft.com/zh-cn/windows/downloads/windows-10-sdk/) + - JDK >= 1.8 [JDK](https://www.oracle.com/cn/java/technologies/javase/javase-jdk8-downloads.html) + + ![project_structure](images/project_structure.png) + + 4.2 NDK版本不匹配问题： + + 打开`Android SDK`，点击`Show Package Details`，根据报错信息选择安装合适的NDK版本。 + ![NDK_version](images/NDK_version.jpg) - ![result](images/object_detection.png) + 4.3 Android Studio版本问题： -4. Android Studio 配置问题解决方案可参考下表： + 在`工具栏-help-Checkout for Updates`中更新Android Studio版本。 - | | 报错 | 解决方案 | - | ---- | ------------------------------------------------------------ | ------------------------------------------------------------ | - | 1 | Gradle sync failed: NDK not configured. | 在local.properties中指定安装的ndk目录：ndk.dir={ndk的安装目录} | - | 2 | Requested NDK version did not match the version requested by ndk.dir | 可手动下载相应的[NDK版本](https://developer.android.com/ndk/downloads?hl=zh-cn)，并在Project Structure - Android NDK location设置中指定SDK的位置（可参考下图完成） | - | 3 | This version of Android Studio cannot open this project, please retry with Android Studio or newer. | 在工具栏-help-Checkout for Updates中更新版本 | - | 4 | SSL peer shut down incorrectly | 重新构建 | + 4.4 Gradle下依赖项安装过慢问题： - ![project_structure](images/project_structure.png) + 如图所示， 打开Demo根目录下`build.gradle`文件，加入华为镜像源地址：`maven {url 'https://developer.huawei.com/repo/'}`，修改classpath为4.0.0，点击`sync`进行同步。下载完成后，将classpath版本复原，再次进行同步。 + ![maven](images/maven.jpg) ## 示例程序详细说明 diff --git a/model_zoo/official/lite/object_detection/images/NDK_version.jpg b/model_zoo/official/lite/object_detection/images/NDK_version.jpg new file mode 100644 index 0000000000..50af23b0db Binary files /dev/null and b/model_zoo/official/lite/object_detection/images/NDK_version.jpg differ diff --git a/model_zoo/official/lite/object_detection/images/maven.jpg b/model_zoo/official/lite/object_detection/images/maven.jpg new file mode 100644 index 0000000000..eee21bb1df Binary files /dev/null and b/model_zoo/official/lite/object_detection/images/maven.jpg differ diff --git a/model_zoo/official/lite/object_detection/images/sdk_management.jpg b/model_zoo/official/lite/object_detection/images/sdk_management.jpg new file mode 100644 index 0000000000..4051b9bab3 Binary files /dev/null and b/model_zoo/official/lite/object_detection/images/sdk_management.jpg differ diff --git a/model_zoo/official/lite/object_detection/images/sdk_management.png b/model_zoo/official/lite/object_detection/images/sdk_management.png deleted file mode 100644 index faf694bd2e..0000000000 Binary files a/model_zoo/official/lite/object_detection/images/sdk_management.png and /dev/null differ diff --git a/model_zoo/official/lite/posenet/README.en.md b/model_zoo/official/lite/posenet/README.en.md index f6e8b585c2..a625e71a0e 100644 --- a/model_zoo/official/lite/posenet/README.en.md +++ b/model_zoo/official/lite/posenet/README.en.md @@ -5,52 +5,65 @@ This sample application demonstrates how to use the MindSpore Lite API and skele ## Running Dependencies - Android Studio 3.2 or later (Android 4.0 or later is recommended.) -- NDK 21.3 -- CMake 3.10 -- Android software development kit (SDK) 26 or later ## Building and Running -1. Load the sample source code to Android Studio and install the corresponding SDK. (After the SDK version is specified, Android Studio automatically installs the SDK.) +1. Load the sample source code to Android Studio. ![start_home](images/home.png) - Start Android Studio, click `File > Settings > System Settings > Android SDK`, and select the corresponding SDK. As shown in the following figure, select an SDK and click `OK`. Android Studio automatically installs the SDK. + Start Android Studio, click `File > Settings > System Settings > Android SDK`, and select the corresponding `SDK Tools`. As shown in the following figure, select an SDK and click `OK`. Android Studio automatically installs the SDK. - ![start_sdk](images/sdk_management.png) + ![start_sdk](images/sdk_management.jpg) - If an Android Studio configuration error occurs, solve it by referring to the following solution table in item 4. + > Android SDK Tools is the default installation. You can see this by unchecking the `Hide Obsolete Packages`box. + > + > If you have any Android Studio configuration problem when trying this demo, please refer to item 4 to resolve it. -2. Connect to an Android device and run the skeleton detection sample application. +2. Connect to an Android device and runs this application. Connect to the Android device through a USB cable for debugging. Click `Run 'app'` to run the sample project on your device. - > During the building, Android Studio automatically downloads dependencies related to MindSpore Lite and model files. Please wait. ![run_app](images/run_app.PNG) - For details about how to connect the Android Studio to a device for debugging, see . + > Android Studio will automatically download MindSpore Lite, model files and other dependencies during the compilation process. Please be patient during this process. + > + > For details about how to connect the Android Studio to a device for debugging, see . + > + > The mobile phone needs to be turn on "USB debugging mode" before Android Studio can recognize the mobile phone. Huawei mobile phones generally turn on "USB debugging model" in Settings -> system and update -> developer Options -> USB debugging. 3. Continue the installation on the Android device. After the installation is complete, you can view the content captured by a camera and the inference result. - ![install](images/install.jpg) + ![result](images/app_result.jpg) - The following figure shows the output of the skeletal detection model. +4. The solutions of configuration problems: - The blue points are used to detect facial features and limb bone movement trends. The confidence score of this inference is 0.98/1, and the inference delay is 66.77 ms. + 4.1 Problems of NDK, CMake, JDK Tools: - ![sult](images/posenet_detection.png) + If the tools installed in Android Studio are not recognized, you can re-download and install them from the corresponding official website, and configure the path. -4. The following table lists solutions to Android Studio configuration errors. - - | | Error | Solution | - | ---- | ------------------------------------------------------------ | ------------------------------------------------------------ | - | 1 | Gradle sync failed: NDK not configured. | Specify the NDK installation directory in the local.properties file: ndk.dir={NDK installation directory} | - | 2 | Requested NDK version did not match the version requested by ndk.dir | Manually download the corresponding [NDK version](https://developer.android.com/ndk/downloads) and specify the SDK location in the `Android NDK location` field (see the following figure). | - | 3 | This version of Android Studio cannot open this project, please retry with Android Studio or newer. | Choose `Help` > `Checkout for Updates` on the toolbar to update the version. | - | 4 | SSL peer shut down incorrectly | Rebuild. | + - NDK >= 21.3 [NDK](https://developer.android.google.cn/ndk/downloads?hl=zh-cn) + - CMake >= 3.10.2 [CMake](https://cmake.org/download) + - Android SDK >= 26 [SDK](https://developer.microsoft.com/zh-cn/windows/downloads/windows-10-sdk/) + - JDK >= 1.8 [JDK](https://www.oracle.com/cn/java/technologies/javase/javase-jdk8-downloads.html) ![project_structure](images/project_structure.png) + 4.2 NDK version does not match: + + Open `Android SDK`, click `Show Package Details`, and select the appropriate NDK version according to the error message. + ![NDK_version](images/NDK_version.jpg) + + 4.3 Problem of Android Studio version: + + Update the Android Studio version in `Toolbar - Help - Checkout for Updates`. + + 4.4 Gradle dependencies installed too slowly: + + As shown in the picture, open the Demo root directory `build. Gradle` file, then add huawei mirror source address: `maven {url 'https://developer.huawei.com/repo/'}`, modify the classpath to 4.0.0 and click ` sync ` . Once the download is complete, restore the classpath version and synchronize it again. + + ![maven](images/maven.jpg) + ## Detailed Description of the Sample Application The skeleton detection sample application on the Android device uses the Android Camera 2 API to enable a camera to obtain image frames and process images, as well as using [runtime](https://www.mindspore.cn/tutorial/lite/en/master/use/runtime.html) to complete model inference. diff --git a/model_zoo/official/lite/posenet/README.md b/model_zoo/official/lite/posenet/README.md index 515f18b3be..17a07ded23 100644 --- a/model_zoo/official/lite/posenet/README.md +++ b/model_zoo/official/lite/posenet/README.md @@ -5,51 +5,67 @@ ## 运行依赖 - Android Studio >= 3.2 (推荐4.0以上版本) -- NDK 21.3 -- CMake 3.10 -- Android SDK >= 26 ## 构建与运行 -1. 在Android Studio中加载本示例源码，并安装相应的SDK（指定SDK版本后，由Android Studio自动安装）。 +1. 在Android Studio中加载本示例源码。 ![start_home](images/home.png) - 启动Android Studio后，点击`File->Settings->System Settings->Android SDK`，勾选相应的SDK。如下图所示，勾选后，点击`OK`，Android Studio即可自动安装SDK。 + 启动Android Studio后，点击`File->Settings->System Settings->Android SDK`，勾选相应的`SDK Tools`。如下图所示，勾选后，点击`OK`，Android Studio即可自动安装SDK。 - ![start_sdk](images/sdk_management.png) + ![start_sdk](images/sdk_management.jpg) - 使用过程中若出现Android Studio配置问题，可参考第4项解决。 + > Android SDK Tools为默认安装项，取消`Hide Obsolete Packages`选框之后可看到。 + > + > 使用过程中若出现问题，可参考第4项解决。 -2. 连接Android设备，运行骨骼检测示例应用程序。 +2. 连接Android设备，运行该应用程序。 + + 通过USB连接Android手机。待成功识别到设备后，点击`Run 'app'`即可在您的手机上运行本示例项目。 - 通过USB连接Android设备调试，点击`Run 'app'`即可在你的设备上运行本示例项目。 > 编译过程中Android Studio会自动下载MindSpore Lite、模型文件等相关依赖项，编译过程需做耐心等待。 + > + > Android Studio连接设备调试操作，可参考。 + > + > 手机需开启“USB调试模式”，Android Studio 才能识别到手机。 华为手机一般在设置->系统和更新->开发人员选项->USB调试中开始“USB调试模型”。 ![run_app](images/run_app.PNG) - Android Studio连接设备调试操作，可参考。 - 3. 在Android设备上，点击“继续安装”，安装完即可查看到设备摄像头捕获的内容和推理结果。 ![install](images/install.jpg) - 使用骨骼检测模型的输出如图： + 如下图所示，识别出的概率最高的物体是植物。 + + ![result](images/app_result.jpg) + +4. Demo部署问题解决方案。 + + 4.1 NDK、CMake、JDK等工具问题： + + 如果Android Studio内安装的工具出现无法识别等问题，可重新从相应官网下载和安装，并配置路径。 + + - NDK >= 21.3 [NDK](https://developer.android.google.cn/ndk/downloads?hl=zh-cn) + - CMake >= 3.10.2 [CMake](https://cmake.org/download) + - Android SDK >= 26 [SDK](https://developer.microsoft.com/zh-cn/windows/downloads/windows-10-sdk/) + - JDK >= 1.8 [JDK](https://www.oracle.com/cn/java/technologies/javase/javase-jdk8-downloads.html) + + ![project_structure](images/project_structure.png) + + 4.2 NDK版本不匹配问题： - 蓝色标识点检测人体面部的五官分布及上肢、下肢的骨骼走势。此次推理置信分数0.98/1，推理时延66.77ms。 + 打开`Android SDK`，点击`Show Package Details`，根据报错信息选择安装合适的NDK版本。 + ![NDK_version](images/NDK_version.jpg) - ![sult](images/posenet_detection.png) + 4.3 Android Studio版本问题： -4. Android Studio 配置问题解决方案可参考下表： + 在`工具栏-help-Checkout for Updates`中更新Android Studio版本。 - | | 报错 | 解决方案 | - | ---- | ------------------------------------------------------------ | ------------------------------------------------------------ | - | 1 | Gradle sync failed: NDK not configured. | 在local.properties中指定安装的ndk目录：ndk.dir={ndk的安装目录} | - | 2 | Requested NDK version did not match the version requested by ndk.dir | 可手动下载相应的[NDK版本](https://developer.android.com/ndk/downloads?hl=zh-cn)，并在Project Structure - Android NDK location设置中指定SDK的位置（可参考下图完成） | - | 3 | This version of Android Studio cannot open this project, please retry with Android Studio or newer. | 在工具栏-help-Checkout for Updates中更新版本 | - | 4 | SSL peer shut down incorrectly | 重新构建 | + 4.4 Gradle下依赖项安装过慢问题： - ![project_structure](images/project_structure.png) + 如图所示， 打开Demo根目录下`build.gradle`文件，加入华为镜像源地址：`maven {url 'https://developer.huawei.com/repo/'}`，修改classpath为4.0.0，点击`sync`进行同步。下载完成后，将classpath版本复原，再次进行同步。 + ![maven](images/maven.jpg) ## 示例程序详细说明 diff --git a/model_zoo/official/lite/posenet/images/NDK_version.jpg b/model_zoo/official/lite/posenet/images/NDK_version.jpg new file mode 100644 index 0000000000..50af23b0db Binary files /dev/null and b/model_zoo/official/lite/posenet/images/NDK_version.jpg differ diff --git a/model_zoo/official/lite/posenet/images/maven.jpg b/model_zoo/official/lite/posenet/images/maven.jpg new file mode 100644 index 0000000000..eee21bb1df Binary files /dev/null and b/model_zoo/official/lite/posenet/images/maven.jpg differ diff --git a/model_zoo/official/lite/posenet/images/sdk_management.jpg b/model_zoo/official/lite/posenet/images/sdk_management.jpg new file mode 100644 index 0000000000..4051b9bab3 Binary files /dev/null and b/model_zoo/official/lite/posenet/images/sdk_management.jpg differ diff --git a/model_zoo/official/lite/posenet/images/sdk_management.png b/model_zoo/official/lite/posenet/images/sdk_management.png deleted file mode 100644 index faf694bd2e..0000000000 Binary files a/model_zoo/official/lite/posenet/images/sdk_management.png and /dev/null differ diff --git a/model_zoo/official/lite/scene_detection/README.en.md b/model_zoo/official/lite/scene_detection/README.en.md index dcbc1b3e72..bf8aab1fd6 100644 --- a/model_zoo/official/lite/scene_detection/README.en.md +++ b/model_zoo/official/lite/scene_detection/README.en.md @@ -5,43 +5,64 @@ This sample application demonstrates how to use the MindSpore Lite C++ API (Andr ## Running Dependencies - Android Studio 3.2 or later (Android 4.0 or later is recommended.) -- NDK 21.3 -- CMake 3.10 -- Android software development kit (SDK) 26 or later ## Building and Running -1. Load the sample source code to Android Studio and install the corresponding SDK. (After the SDK version is specified, Android Studio automatically installs the SDK.) +1. Load the sample source code to Android Studio. ![start_home](images/home.png) - Start Android Studio, click `File > Settings > System Settings > Android SDK`, and select the corresponding SDK. As shown in the following figure, select an SDK and click `OK`. Android Studio automatically installs the SDK. + Start Android Studio, click `File > Settings > System Settings > Android SDK`, and select the corresponding `SDK Tools`. As shown in the following figure, select an SDK and click `OK`. Android Studio automatically installs the SDK. - ![start_sdk](images/sdk_management.png) + ![start_sdk](images/sdk_management.jpg) - If an Android Studio configuration error occurs, solve it by referring to the following table. + > Android SDK Tools is the default installation. You can see this by unchecking the `Hide Obsolete Packages`box. + > + > If you have any Android Studio configuration problem when trying this demo, please refer to item 4 to resolve it. - | | Error | Solution | - | ---- | ------------------------------------------------------------ | ------------------------------------------------------------ | - | 1 | Gradle sync failed: NDK not configured. | Specify the NDK installation directory in the local.properties file: ndk.dir={NDK installation directory} | - | 2 | Requested NDK version did not match the version requested by ndk.dir | Manually download the corresponding [NDK version](https://developer.android.com/ndk/downloads) and specify the SDK location in the `Android NDK location` field (see the following figure). | - | 3 | This version of Android Studio cannot open this project, please retry with Android Studio or newer. | Choose `Help` > `Checkout for Updates` on the toolbar to update the version. | - | 4 | SSL peer shut down incorrectly | Rebuild. | - - ![project_structure](images/project_structure.png) - -2. Connect to an Android device and run the scene detection sample application. +2. Connect to an Android device and runs this application. Connect to the Android device through a USB cable for debugging. Click `Run 'app'` to run the sample project on your device. - > During the building, Android Studio automatically downloads dependencies related to MindSpore Lite and model files. Please wait. ![run_app](images/run_app.PNG) - For details about how to connect the Android Studio to a device for debugging, see . + > Android Studio will automatically download MindSpore Lite, model files and other dependencies during the compilation process. Please be patient during this process. + > + > For details about how to connect the Android Studio to a device for debugging, see . + > + > The mobile phone needs to be turn on "USB debugging mode" before Android Studio can recognize the mobile phone. Huawei mobile phones generally turn on "USB debugging model" in Settings -> system and update -> developer Options -> USB debugging. 3. Continue the installation on the Android device. After the installation is complete, you can view the content captured by a camera and the inference result. - ![install](images/install.jpg) + ![result](images/app_result.jpg) + +4. The solutions of configuration problems: + + 4.1 Problems of NDK, CMake, JDK Tools: + + If the tools installed in Android Studio are not recognized, you can re-download and install them from the corresponding official website, and configure the path. + + - NDK >= 21.3 [NDK](https://developer.android.google.cn/ndk/downloads?hl=zh-cn) + - CMake >= 3.10.2 [CMake](https://cmake.org/download) + - Android SDK >= 26 [SDK](https://developer.microsoft.com/zh-cn/windows/downloads/windows-10-sdk/) + - JDK >= 1.8 [JDK](https://www.oracle.com/cn/java/technologies/javase/javase-jdk8-downloads.html) + + ![project_structure](images/project_structure.png) + + 4.2 NDK version does not match: + + Open `Android SDK`, click `Show Package Details`, and select the appropriate NDK version according to the error message. + ![NDK_version](images/NDK_version.jpg) + + 4.3 Problem of Android Studio version: + + Update the Android Studio version in `Toolbar - Help - Checkout for Updates`. + + 4.4 Gradle dependencies installed too slowly: + + As shown in the picture, open the Demo root directory `build. Gradle` file, then add huawei mirror source address: `maven {url 'https://developer.huawei.com/repo/'}`, modify the classpath to 4.0.0 and click ` sync ` . Once the download is complete, restore the classpath version and synchronize it again. + + ![maven](images/maven.jpg) ## Detailed Description of the Sample Application diff --git a/model_zoo/official/lite/scene_detection/README.md b/model_zoo/official/lite/scene_detection/README.md index 77f988d583..aca2e0fb54 100644 --- a/model_zoo/official/lite/scene_detection/README.md +++ b/model_zoo/official/lite/scene_detection/README.md @@ -1,53 +1,73 @@ -## MindSpore Lite 端侧场景检测demo（Android） +# MindSpore Lite 端侧场景检测demo（Android） 本示例程序演示了如何在端侧利用MindSpore Lite C++ API（Android JNI）以及MindSpore Lite 场景检测模型完成端侧推理，对设备摄像头捕获的内容进行检测，并在App图像预览界面中显示连续目标检测结果。 -### 运行依赖 +## 运行依赖 - Android Studio >= 3.2 (推荐4.0以上版本) -- NDK 21.3 -- CMake 3.10 -- Android SDK >= 26 -### 构建与运行 +## 构建与运行 -1. 在Android Studio中加载本示例源码，并安装相应的SDK（指定SDK版本后，由Android Studio自动安装）。 +1. 在Android Studio中加载本示例源码。 ![start_home](images/home.png) - 启动Android Studio后，点击`File->Settings->System Settings->Android SDK`，勾选相应的SDK。如下图所示，勾选后，点击`OK`，Android Studio即可自动安装SDK。 + 启动Android Studio后，点击`File->Settings->System Settings->Android SDK`，勾选相应的`SDK Tools`。如下图所示，勾选后，点击`OK`，Android Studio即可自动安装SDK。 - ![start_sdk](images/sdk_management.png) + ![start_sdk](images/sdk_management.jpg) - 使用过程中若出现Android Studio配置问题，可参考下表解决： + > Android SDK Tools为默认安装项，取消`Hide Obsolete Packages`选框之后可看到。 + > + > 使用过程中若出现问题，可参考第4项解决。 - | | 报错 | 解决方案 | - | ---- | ------------------------------------------------------------ | ------------------------------------------------------------ | - | 1 | Gradle sync failed: NDK not configured. | 在local.properties中指定安装的ndk目录：ndk.dir={ndk的安装目录} | - | 2 | Requested NDK version did not match the version requested by ndk.dir | 可手动下载相应的[NDK版本](https://developer.android.com/ndk/downloads?hl=zh-cn)，并在Project Structure - Android NDK location设置中指定SDK的位置（可参考下图完成） | - | 3 | This version of Android Studio cannot open this project, please retry with Android Studio or newer. | 在工具栏-help-Checkout for Updates中更新版本 | - | 4 | SSL peer shut down incorrectly | 重新构建 | +2. 连接Android设备，运行该应用程序。 - ![project_structure](images/project_structure.png) + 通过USB连接Android手机。待成功识别到设备后，点击`Run 'app'`即可在您的手机上运行本示例项目。 -2. 连接Android设备，运行场景检测示例应用程序。 - - 通过USB连接Android设备调试，点击`Run 'app'`即可在你的设备上运行本示例项目。 - > 编译过程中Android Studio会自动下载MindSpore Lite、模型文件等相关依赖项，编译过程需做耐心等待。 + > 编译过程中Android Studio会自动下载MindSpore Lite、模型文件等相关依赖项，编译过程需做耐心等待。 + > + > Android Studio连接设备调试操作，可参考。 + > + > 手机需开启“USB调试模式”，Android Studio 才能识别到手机。 华为手机一般在设置->系统和更新->开发人员选项->USB调试中开始“USB调试模型”。 ![run_app](images/run_app.PNG) - Android Studio连接设备调试操作，可参考。 - -3. 在Android设备上，点击“继续安装”。完成之后即可在手机上体验场景检测功能。 +3. 在Android设备上，点击“继续安装”，安装完即可查看到设备摄像头捕获的内容和推理结果。 ![install](images/install.jpg) -## 示例程序详细说明 + 如下图所示，识别出的概率最高的物体是植物。 + + ![result](images/app_result.jpg) + +4. Demo部署问题解决方案。 + + 4.1 NDK、CMake、JDK等工具问题： + + 如果Android Studio内安装的工具出现无法识别等问题，可重新从相应官网下载和安装，并配置路径。 + + - NDK >= 21.3 [NDK](https://developer.android.google.cn/ndk/downloads?hl=zh-cn) + - CMake >= 3.10.2 [CMake](https://cmake.org/download) + - Android SDK >= 26 [SDK](https://developer.microsoft.com/zh-cn/windows/downloads/windows-10-sdk/) + - JDK >= 1.8 [JDK](https://www.oracle.com/cn/java/technologies/javase/javase-jdk8-downloads.html) -端侧场景检测Android示例程序分为JAVA层和JNI层，其中，JAVA层主要通过Android Camera 2 API实现摄像头获取图像帧，以及相应的图像处理（针对推理结果画框）等功能；JNI层在[Runtime](https://www.mindspore.cn/tutorial/lite/zh-CN/master/use/runtime.html)中完成模型推理的过程。 + ![project_structure](images/project_structure.png) -> 此处详细说明示例程序的JNI层实现，JAVA层运用Android Camera 2 API实现开启设备摄像头以及图像帧处理等功能，需读者具备一定的Android开发基础知识。 + 4.2 NDK版本不匹配问题： + + 打开`Android SDK`，点击`Show Package Details`，根据报错信息选择安装合适的NDK版本。 + ![NDK_version](images/NDK_version.jpg) + + 4.3 Android Studio版本问题： + + 在`工具栏-help-Checkout for Updates`中更新Android Studio版本。 + + 4.4 Gradle下依赖项安装过慢问题： + + 如图所示， 打开Demo根目录下`build.gradle`文件，加入华为镜像源地址：`maven {url 'https://developer.huawei.com/repo/'}`，修改classpath为4.0.0，点击`sync`进行同步。下载完成后，将classpath版本复原，再次进行同步。 + ![maven](images/maven.jpg) + +## 示例程序详细说明 ### 示例程序结构 diff --git a/model_zoo/official/lite/scene_detection/images/NDK_version.jpg b/model_zoo/official/lite/scene_detection/images/NDK_version.jpg new file mode 100644 index 0000000000..50af23b0db Binary files /dev/null and b/model_zoo/official/lite/scene_detection/images/NDK_version.jpg differ diff --git a/model_zoo/official/lite/scene_detection/images/maven.jpg b/model_zoo/official/lite/scene_detection/images/maven.jpg new file mode 100644 index 0000000000..eee21bb1df Binary files /dev/null and b/model_zoo/official/lite/scene_detection/images/maven.jpg differ diff --git a/model_zoo/official/lite/scene_detection/images/sdk_management.jpg b/model_zoo/official/lite/scene_detection/images/sdk_management.jpg new file mode 100644 index 0000000000..4051b9bab3 Binary files /dev/null and b/model_zoo/official/lite/scene_detection/images/sdk_management.jpg differ diff --git a/model_zoo/official/lite/scene_detection/images/sdk_management.png b/model_zoo/official/lite/scene_detection/images/sdk_management.png deleted file mode 100644 index faf694bd2e..0000000000 Binary files a/model_zoo/official/lite/scene_detection/images/sdk_management.png and /dev/null differ diff --git a/model_zoo/official/lite/style_transfer/README.en.md b/model_zoo/official/lite/style_transfer/README.en.md index 68f9add1ac..aa362e41d3 100644 --- a/model_zoo/official/lite/style_transfer/README.en.md +++ b/model_zoo/official/lite/style_transfer/README.en.md @@ -5,55 +5,64 @@ This sample application demonstrates how to use the MindSpore Lite API and MindS ## Running Dependencies - Android Studio 3.2 or later (Android 4.0 or later is recommended.) -- NDK 21.3 -- CMake 3.10 -- Android software development kit (SDK) 26 or later ## Building and Running -1. Load the sample source code to Android Studio and install the corresponding SDK. (After the SDK version is specified, Android Studio automatically installs the SDK.) +1. Load the sample source code to Android Studio. ![start_home](images/home.png) - Start Android Studio, click `File > Settings > System Settings > Android SDK`, and select the corresponding SDK. As shown in the following figure, select an SDK and click `OK`. Android Studio automatically installs the SDK. + Start Android Studio, click `File > Settings > System Settings > Android SDK`, and select the corresponding `SDK Tools`. As shown in the following figure, select an SDK and click `OK`. Android Studio automatically installs the SDK. - ![start_sdk](images/sdk_management.png) + ![start_sdk](images/sdk_management.jpg) - If an Android Studio configuration error occurs, solve it by referring to the following solution table in item 4. + > Android SDK Tools is the default installation. You can see this by unchecking the `Hide Obsolete Packages`box. + > + > If you have any Android Studio configuration problem when trying this demo, please refer to item 4 to resolve it. -2. Connect to an Android device and run the style transfer sample application. +2. Connect to an Android device and runs this application. Connect to the Android device through a USB cable for debugging. Click `Run 'app'` to run the sample project on your device. - > During the building, Android Studio automatically downloads dependencies related to MindSpore Lite and model files. Please wait. ![run_app](images/run_app.PNG) - For details about how to connect the Android Studio to a device for debugging, see . + > Android Studio will automatically download MindSpore Lite, model files and other dependencies during the compilation process. Please be patient during this process. + > + > For details about how to connect the Android Studio to a device for debugging, see . + > + > The mobile phone needs to be turn on "USB debugging mode" before Android Studio can recognize the mobile phone. Huawei mobile phones generally turn on "USB debugging model" in Settings -> system and update -> developer Options -> USB debugging. -3. Continue the installation on the Android device. After the installation is complete, you can view the inference result. +3. Continue the installation on the Android device. After the installation is complete, you can view the content captured by a camera and the inference result. - ![install](images/install.jpg) + ![result](images/app_result.jpg) - When using the style transfer demo, you can import or take a photo, select a built-in style to obtain a new photo after inference, and then restore or save the new photo. +4. The solutions of configuration problems: - Before style transfer: + 4.1 Problems of NDK, CMake, JDK Tools: - ![sult](images/style_transfer_demo.png) + If the tools installed in Android Studio are not recognized, you can re-download and install them from the corresponding official website, and configure the path. - After style transfer: + - NDK >= 21.3 [NDK](https://developer.android.google.cn/ndk/downloads?hl=zh-cn) + - CMake >= 3.10.2 [CMake](https://cmake.org/download) + - Android SDK >= 26 [SDK](https://developer.microsoft.com/zh-cn/windows/downloads/windows-10-sdk/) + - JDK >= 1.8 [JDK](https://www.oracle.com/cn/java/technologies/javase/javase-jdk8-downloads.html) - ![sult](images/style_transfer_result.png) + ![project_structure](images/project_structure.png) -4. The following table lists solutions to Android Studio configuration errors. + 4.2 NDK version does not match: - | | Error | Solution | - | ---- | ------------------------------------------------------------ | ------------------------------------------------------------ | - | 1 | Gradle sync failed: NDK not configured. | Specify the NDK installation directory in the local.properties file: ndk.dir={NDK installation directory} | - | 2 | Requested NDK version did not match the version requested by ndk.dir | Manually download the corresponding [NDK version](https://developer.android.com/ndk/downloads) and specify the SDK location in the `Android NDK location` field (see the following figure). | - | 3 | This version of Android Studio cannot open this project, please retry with Android Studio or newer. | Choose `Help` > `Checkout for Updates` on the toolbar to update the version. | - | 4 | SSL peer shut down incorrectly | Rebuild. | + Open `Android SDK`, click `Show Package Details`, and select the appropriate NDK version according to the error message. + ![NDK_version](images/NDK_version.jpg) - ![project_structure](images/project_structure.png) + 4.3 Problem of Android Studio version: + + Update the Android Studio version in `Toolbar - Help - Checkout for Updates`. + + 4.4 Gradle dependencies installed too slowly: + + As shown in the picture, open the Demo root directory `build. Gradle` file, then add huawei mirror source address: `maven {url 'https://developer.huawei.com/repo/'}`, modify the classpath to 4.0.0 and click ` sync ` . Once the download is complete, restore the classpath version and synchronize it again. + + ![maven](images/maven.jpg) ## Detailed Description of the Sample Application @@ -88,7 +97,7 @@ The style transfer sample application on the Android device uses the Android Cam ### Downloading and Deploying the Model File -Download the model file from MindSpore Model Hub. The objective detection model file used in this sample application is `posenet_model.ms`, which is automatically downloaded during app building using the `download.gradle` script and stored in the `app/src/main/assets` project directory. +Download the model file from MindSpore Model Hub. The objective detection model file used in this sample application is `style_predict_quant.ms` and `style_transfer_quant.ms`, which is automatically downloaded during app building using the `download.gradle` script and stored in the `app/src/main/assets` project directory. > If the download fails, manually download the model files [style_predict_quant.ms](https://download.mindspore.cn/model_zoo/official/lite/style_lite/style_predict_quant.ms) and [style_transfer_quant.ms](https://download.mindspore.cn/model_zoo/official/lite/style_lite/style_transfer_quant.ms). diff --git a/model_zoo/official/lite/style_transfer/README.md b/model_zoo/official/lite/style_transfer/README.md index b257ad1b8a..ca1aa9c0f5 100644 --- a/model_zoo/official/lite/style_transfer/README.md +++ b/model_zoo/official/lite/style_transfer/README.md @@ -5,55 +5,67 @@ ## 运行依赖 - Android Studio >= 3.2 (推荐4.0以上版本) -- NDK 21.3 -- CMake 3.10 -- Android SDK >= 26 ## 构建与运行 -1. 在Android Studio中加载本示例源码，并安装相应的SDK（指定SDK版本后，由Android Studio自动安装）。 +1. 在Android Studio中加载本示例源码。 ![start_home](images/home.png) - 启动Android Studio后，点击`File->Settings->System Settings->Android SDK`，勾选相应的SDK。如下图所示，勾选后，点击`OK`，Android Studio即可自动安装SDK。 + 启动Android Studio后，点击`File->Settings->System Settings->Android SDK`，勾选相应的`SDK Tools`。如下图所示，勾选后，点击`OK`，Android Studio即可自动安装SDK。 - ![start_sdk](images/sdk_management.png) + ![start_sdk](images/sdk_management.jpg) - 使用过程中若出现Android Studio配置问题，可参考第4项解决。 + > Android SDK Tools为默认安装项，取消`Hide Obsolete Packages`选框之后可看到。 + > + > 使用过程中若出现问题，可参考第4项解决。 -2. 连接Android设备，运行应用程序。 +2. 连接Android设备，运行该应用程序。 + + 通过USB连接Android手机。待成功识别到设备后，点击`Run 'app'`即可在您的手机上运行本示例项目。 - 通过USB连接Android设备调试，点击`Run 'app'`即可在你的设备上运行本示例项目。 > 编译过程中Android Studio会自动下载MindSpore Lite、模型文件等相关依赖项，编译过程需做耐心等待。 + > + > Android Studio连接设备调试操作，可参考。 + > + > 手机需开启“USB调试模式”，Android Studio 才能识别到手机。 华为手机一般在设置->系统和更新->开发人员选项->USB调试中开始“USB调试模型”。 ![run_app](images/run_app.PNG) - Android Studio连接设备调试操作，可参考。 - -3. 在Android设备上，点击“继续安装”，安装完即可查看到推理结果。 +3. 在Android设备上，点击“继续安装”，安装完即可查看到设备摄像头捕获的内容和推理结果。 ![install](images/install.jpg) - 使用风格迁移demo时，用户可先导入或拍摄自己的图片，然后选择一种预置风格，得到推理后的新图片，最后使用还原或保存新图片功能。 + 如下图所示，识别出的概率最高的物体是植物。 + + ![result](images/app_result.jpg) + +4. Demo部署问题解决方案。 + + 4.1 NDK、CMake、JDK等工具问题： + + 如果Android Studio内安装的工具出现无法识别等问题，可重新从相应官网下载和安装，并配置路径。 + + - NDK >= 21.3 [NDK](https://developer.android.google.cn/ndk/downloads?hl=zh-cn) + - CMake >= 3.10.2 [CMake](https://cmake.org/download) + - Android SDK >= 26 [SDK](https://developer.microsoft.com/zh-cn/windows/downloads/windows-10-sdk/) + - JDK >= 1.8 [JDK](https://www.oracle.com/cn/java/technologies/javase/javase-jdk8-downloads.html) - 原始图片： + ![project_structure](images/project_structure.png) - ![sult](images/style_transfer_demo.png) + 4.2 NDK版本不匹配问题： - 风格迁移后的新图片： + 打开`Android SDK`，点击`Show Package Details`，根据报错信息选择安装合适的NDK版本。 + ![NDK_version](images/NDK_version.jpg) - ![sult](images/style_transfer_result.png) + 4.3 Android Studio版本问题： -4. Android Studio 配置问题解决方案可参考下表： + 在`工具栏-help-Checkout for Updates`中更新Android Studio版本。 - | | 报错 | 解决方案 | - | ---- | ------------------------------------------------------------ | ------------------------------------------------------------ | - | 1 | Gradle sync failed: NDK not configured. | 在local.properties中指定安装的ndk目录：ndk.dir={ndk的安装目录} | - | 2 | Requested NDK version did not match the version requested by ndk.dir | 可手动下载相应的[NDK版本](https://developer.android.com/ndk/downloads?hl=zh-cn)，并在Project Structure - Android NDK location设置中指定SDK的位置（可参考下图完成） | - | 3 | This version of Android Studio cannot open this project, please retry with Android Studio or newer. | 在工具栏-help-Checkout for Updates中更新版本 | - | 4 | SSL peer shut down incorrectly | 重新构建 | + 4.4 Gradle下依赖项安装过慢问题： - ![project_structure](images/project_structure.png) + 如图所示， 打开Demo根目录下`build.gradle`文件，加入华为镜像源地址：`maven {url 'https://developer.huawei.com/repo/'}`，修改classpath为4.0.0，点击`sync`进行同步。下载完成后，将classpath版本复原，再次进行同步。 + ![maven](images/maven.jpg) ## 示例程序详细说明 @@ -88,7 +100,7 @@ ### 下载及部署模型文件 -从MindSpore Model Hub中下载模型文件，本示例程序中使用的目标检测模型文件为`posenet_model.ms`，同样通过`download.gradle`脚本在APP构建时自动下载，并放置在`app/src/main/assets`工程目录下。 +从MindSpore Model Hub中下载模型文件，本示例程序中使用的目标检测模型文件为`style_predict_quant.ms`、`style_transfer_quant.ms`，同样通过`download.gradle`脚本在APP构建时自动下载，并放置在`app/src/main/assets`工程目录下。 > 若下载失败请手动下载模型文件，style_predict_quant.ms [下载链接](https://download.mindspore.cn/model_zoo/official/lite/style_lite/style_predict_quant.ms)，以及style_transfer_quant.ms [下载链接](https://download.mindspore.cn/model_zoo/official/lite/style_lite/style_transfer_quant.ms)。 diff --git a/model_zoo/official/lite/style_transfer/images/NDK_version.jpg b/model_zoo/official/lite/style_transfer/images/NDK_version.jpg new file mode 100644 index 0000000000..50af23b0db Binary files /dev/null and b/model_zoo/official/lite/style_transfer/images/NDK_version.jpg differ diff --git a/model_zoo/official/lite/style_transfer/images/maven.jpg b/model_zoo/official/lite/style_transfer/images/maven.jpg new file mode 100644 index 0000000000..eee21bb1df Binary files /dev/null and b/model_zoo/official/lite/style_transfer/images/maven.jpg differ diff --git a/model_zoo/official/lite/style_transfer/images/sdk_management.jpg b/model_zoo/official/lite/style_transfer/images/sdk_management.jpg new file mode 100644 index 0000000000..4051b9bab3 Binary files /dev/null and b/model_zoo/official/lite/style_transfer/images/sdk_management.jpg differ diff --git a/model_zoo/official/lite/style_transfer/images/sdk_management.png b/model_zoo/official/lite/style_transfer/images/sdk_management.png deleted file mode 100644 index faf694bd2e..0000000000 Binary files a/model_zoo/official/lite/style_transfer/images/sdk_management.png and /dev/null differ