Integrate Inference Engine¶

Integrate Inference Engine with Your C++ Application¶

C++

The following diagram illustrates the typical Inference Engine С++ API workflow:

Read the sections below to learn about each item.

Note

Before start using Inference Engine, make sure you set all environment variables during the installation. If you did not, follow the instructions from the Set the Environment Variables section in the installation guides:

For Windows* 10
For Linux*
For macOS*
To build an open source version, use the Inference Engine Build Instructions.

Link with Inference Library¶

Create a structure for the project:

project/
    ├── CMakeLists.txt  - CMake file to build
    ├── ...             - Additional folders like includes/
    └── src/            - source folder
        └── main.cpp
build/                  - build directory
    ...

Include Inference Engine, nGraph and OpenCV libraries in project/CMakeLists.txt

OpenCV integration is needed mostly for pre-processing input data and nGraph for more complex applications using nGraph API.

cmake_minimum_required(VERSION 3.0.0)
project(project_name)
find_package(ngraph REQUIRED)
find_package(InferenceEngine REQUIRED)
find_package(OpenCV REQUIRED)
add_executable(${PROJECT_NAME} src/main.cpp)
target_link_libraries(${PROJECT_NAME} PRIVATE ${InferenceEngine_LIBRARIES} ${OpenCV_LIBS}    ${NGRAPH_LIBRARIES})

Use Inference Engine API to Implement Inference Pipeline¶

This section provides step-by-step instructions to implement a typical inference pipeline with the Inference Engine C++ API:

_images/IMPLEMENT_PIPELINE_with_API_C.svg

Step 1. Create Inference Engine Core¶

Use the following code to create Inference Engine Core to manage available devices and read network objects:

InferenceEngine::Core core;
InferenceEngine::CNNNetwork network;
InferenceEngine::ExecutableNetwork executable_network;

Step 2 (Optional). Configure Input and Output of the Model¶

Optionally, configure input and output of the model using the steps below:

Load a model to a Core object:

auto network  = core.ReadNetwork("model.xml");

ONNX

auto network = core.ReadNetwork("model.onnx");

You can find more information about the ONNX format support in the document ONNX format support in the OpenVINO™

nGraph

std::shared_ptr<Function> createNetwork() {
   // To construct a network, please follow
   // https://docs.openvinotoolkit.org/latest/openvino_docs_nGraph_DG_build_function.html
}
auto network = CNNNetwork(createNetwork());

Request input and output information using InferenceEngine::CNNNetwork::getInputsInfo(), and InferenceEngine::CNNNetwork::getOutputsInfo() methods:
```
/** Take information about all topology inputs **/
InferenceEngine::InputsDataMap input_info = network.getInputsInfo();
/** Iterate over all input info**/
for (auto &item : input_info) {
    auto input_data = item.second;
        // Add your input configuration steps here
}

/** Take information about all topology outputs **/
InferenceEngine::OutputsDataMap output_info = network.getOutputsInfo();
/** Iterate over all output info**/
for (auto &item : output_info) {
    auto output_data = item.second;
        // Add your output configuration steps here
}
```
Configuring options:
1. Set precision (number format): FP16, FP32, INT8, etc. Refer to the Supported Configurations section on the Supported Devices page to choose the relevant configuration.
  
  For input (iterate over all input info):
```
input_data->setPrecision(InferenceEngine::Precision::U8);
```
  For output (iterate over all output info):
```
output_data->setPrecision(InferenceEngine::Precision::FP32);
```
  By default, the input and output precision is set to Precision::FP32.
2. Set layout (NCHW, ).
  
  For input (iterate over all input info):
```
input_data->setLayout(InferenceEngine::Layout::NCHW);
```
  By default, the input layout is set to Layout::NCHW.
  
  For output (iterate over all output info):
```
output_data->setLayout(InferenceEngine::Layout::NC);
```
  By default, the output layout depends on a number of its dimensions:
  
  Number of dimensions
  
  5
  
  4
  
  3
  
  2
  
  1
  
  Layout
  
  NCDHW
  
  NCHW
  
  CHW
  
  NC
  
  C
3. Set resize algorithm for inputs (Bilinear). You can allow input of any size. To do this, mark each input as resizable by setting a desired resize algorithm (e.g. BILINEAR) inside of the appropriate input info (Iterate over all input info):
```
input_data->getPreProcess().setResizeAlgorithm(InferenceEngine::RESIZE_BILINEAR);
```
  By default, no resize algorithm is set for inputs.
4. Set color format (BGR, RGB, NV12). Basic color format conversions are supported as well. By default, the Inference Engine assumes that the input color format is BGR and color format conversions are disabled. Set ColorFormat::RAW input color format if the input does not need color conversions. The Inference Engine supports the following color format conversions:
  - RGB->BGR
  - RGBX->BGR
  - BGRX->BGR
  - NV12->BGR where X is a channel that will be ignored during inference. To enable the conversions, set a desired color format (for example, RGB) for each input inside of the appropriate input info (iterate over all input info):
    input_data->getPreProcess().setColorFormat(InferenceEngine::ColorFormat::RGB);
    Note
    
    NV12 input color format pre-processing differs from other color conversions. In case of NV12, Inference Engine expects two separate image planes (Y and UV). You must use a specific InferenceEngine::NV12Blob object instead of default blob object and set this blob to the Inference Engine Infer Request using InferenceEngine::InferRequest::SetBlob(). Refer to Hello NV12 Input Classification C++ Sample for more details.
5. Run on multiple images with setting batch. If you want to run inference for multiple images at once, you can use the built-in batch pre-processing functionality.
  
  NOTE : Batch pre-processing is not supported if input color format is set to ColorFormat::NV12.

Number of dimensions	5	4	3	2	1
Layout	NCDHW	NCHW	CHW	NC	C

Step 3. Load the Model to the Device¶

Load the model to the device using InferenceEngine::Core::LoadNetwork() :

executable_network = core.LoadNetwork("model.xml", "CPU");

ONNX

executable_network = core.LoadNetwork("model.onnx", "CPU");

nGraph

std::shared_ptr<Function> createNetwork() {
   // To construct a network, please follow
   // https://docs.openvinotoolkit.org/latest/openvino_docs_nGraph_DG_build_function.html
}
auto network = CNNNetwork(createNetwork());
executable_network = core.LoadNetwork(network, "CPU");

Model From Step 2

Follow this step only if you went through optional “Step 2 (Optional). Configure Input and Output of the Model”, otherwise use another tab for your model type: IR (OpenVINO Intermediate Representation), ONNX or nGraph.

executable_network = core.LoadNetwork(network, "CPU");

It creates an executable network from a network object. The executable network is associated with single hardware device. It is possible to create as many networks as needed and to use them simultaneously (up to the limitation of the hardware resources).

Third parameter is a configuration for plugin. It is map of pairs: (parameter name, parameter value). Choose device from Supported devices page for more details about supported configuration parameters.

/** Optional config. E.g. this enables profiling of performance counters. **/
std::map<std::string, std::string> config = {{ InferenceEngine::PluginConfigParams::KEY_PERF_COUNT, InferenceEngine::PluginConfigParams::YES }};
executable_network = core.LoadNetwork(network, "CPU", config);

Step 4. Create an Inference Request¶

Create an infer request using the following code:

auto infer_request = executable_network.CreateInferRequest();

Step 5. Prepare Input¶

You can use one of the following options to prepare input:

Optimal way for a single network. Get blobs allocated by an infer request using InferenceEngine::InferRequest::GetBlob() and feed an image and the input data to the blobs. In this case, input data must be aligned (resized manually) with a given blob size and have a correct color format.

/** Iterate over all input blobs **/
for (auto & item : input_info) {
    auto input_name = item.first;
    /** Get input blob **/
    auto input = infer_request.GetBlob(input_name);
    /** Fill input tensor with planes. First b channel, then g and r channels **/
//     ...
}

Optimal way for a cascade of networks (output of one network is input for another). Get output blob from the first request using InferenceEngine::InferRequest::GetBlob() and set it as input for the second request using InferenceEngine::InferRequest::SetBlob().
```
auto output = infer_request1.GetBlob(output_name);
infer_request2.SetBlob(input_name, output);
```
Optimal way to handle ROI (a ROI object located inside of input of one network is input for another). It is possible to re-use shared input by several networks. You do not need to allocate separate input blob for a network if it processes a ROI object located inside of already allocated input of a previous network. For instance, when first network detects objects on a video frame (stored as input blob) and second network accepts detected bounding boxes (ROI inside of the frame) as input. In this case, it is allowed to re-use pre-allocated input blob (used by first network) by second network and just crop ROI without allocation of new memory using InferenceEngine::make_shared_blob() with passing of InferenceEngine::Blob::Ptr and InferenceEngine::ROI as parameters.
```
/** inputBlob points to input of a previous network and
    cropROI contains coordinates of output bounding box **/
InferenceEngine::Blob::Ptr inputBlob;
InferenceEngine::ROI cropRoi;
//...

/** roiBlob uses shared memory of inputBlob and describes cropROI
    according to its coordinates **/
auto roiBlob = InferenceEngine::make_shared_blob(inputBlob, cropRoi);
infer_request2.SetBlob(input_name, roiBlob);
```
Make sure that shared input is kept valid during execution of each network. Otherwise, ROI blob may be corrupted if the original input blob (that ROI is cropped from) has already been rewritten.

Allocate input blobs of the appropriate types and sizes, feed an image and the input data to the blobs, and call InferenceEngine::InferRequest::SetBlob() to set these blobs for an infer request:

/** Iterate over all input blobs **/
for (auto & item : input_info) {
    auto input_data = item.second;
    /** Create input blob **/
    InferenceEngine::TBlob<unsigned char>::Ptr input;
    // assuming input precision was asked to be U8 in prev step
    input = InferenceEngine::make_shared_blob<unsigned char>(
        InferenceEngine::TensorDesc(InferenceEngine::Precision::U8, input_data->getTensorDesc().getDims(),
        input_data->getTensorDesc().getLayout()));
    input->allocate();
    infer_request.SetBlob(item.first, input);

    /** Fill input tensor with planes. First b channel, then g and r channels **/
//     ...
}

A blob can be filled before and after SetBlob().

Note

The SetBlob() method compares precision and layout of an input blob with the ones defined in step 3 and throws an exception if they do not match. It also compares a size of the input blob with input size of the read network. But if input was configured as resizable, you can set an input blob of any size (for example, any ROI blob). Input resize will be invoked automatically using resize algorithm configured on step 3. Similarly to the resize, color format conversions allow the color format of an input blob to differ from the color format of the read network. Color format conversion will be invoked automatically using color format configured on step 3.

GetBlob() logic is the same for pre-processable and not pre-processable input. Even if it is called with input configured as resizable or as having specific color format, a blob allocated by an infer request is returned. Its size and color format are already consistent with the corresponding values of the read network. No pre-processing will happen for this blob. If you call GetBlob() after SetBlob(), you will get the blob you set in SetBlob().

Step 6. Start Inference¶

Start inference in asynchronous or synchronous mode. Async API usage can improve overall frame-rate of the application, because rather than wait for inference to complete, the app can continue doing things on the host, while accelerator is busy.

For synchronous inference request:
```
infer_request.Infer();
```
For asynchronous inference request:
```
infer_request.StartAsync();
infer_request.Wait(InferenceEngine::InferRequest::WaitMode::RESULT_READY);
```
StartAsync returns immediately and starts inference without blocking main thread, Infer blocks main thread and returns when inference is completed. Call Wait for waiting result to become available for asynchronous request.

There are three ways to use it:
- specify maximum duration in milliseconds to block for. The method is blocked until the specified timeout has elapsed, or the result becomes available, whichever comes first.
- InferenceEngine::InferRequest::WaitMode::RESULT_READY - waits until inference result becomes available
- InferenceEngine::InferRequest::WaitMode::STATUS_ONLY - immediately returns request status.It does not block or interrupts current thread.

Both requests are thread-safe: can be called from different threads without fearing corruption and failures.

Multiple requests for single ExecutableNetwork are executed sequentially one by one in FIFO order.

While request is ongoing, all its methods except InferenceEngine::InferRequest::Wait would throw an exception.

Step 7. Process the Inference Results¶

Go over the output blobs and process the inference results. Note that casting Blob to TBlob via std::dynamic_pointer_cast is not the recommended way. It’s better to access data via the buffer() and as() methods as follows:

for (auto &item : output_info) {
    auto output_name = item.first;
    auto output = infer_request.GetBlob(output_name);
    {
        auto const memLocker = output->cbuffer(); // use const memory locker
        // output_buffer is valid as long as the lifetime of memLocker
        const float *output_buffer = memLocker.as<const float *>();
        /** output_buffer[] - accessing output blob data **/
    }
}

Build Your Application¶

For details about building your application, refer to the CMake files for the sample applications. All samples source code is located in the <INSTALL_DIR>/openvino/inference_engine/samples directory, where INSTALL_DIR is the OpenVINO™ installation directory.

To build your project using CMake with the default build tools currently available on your machine, execute the following commands:

Note

Make sure you set environment variables first by running <INSTALL_DIR>/bin/setupvars.sh (or setupvars.bat for Windows). Otherwise the InferenceEngine_DIR and OpenCV_DIR variables won’t be configured properly to pass find_package calls.

cd build/
cmake ../project
cmake --build .

It’s allowed to specify additional build options (e.g. to build CMake project on Windows with a specific build tools). Please refer to the CMake page for details.

Run Your Application¶

Note

Before running, make sure you completed Set the Environment Variables section in OpenVINO Installation document so that the application can find the libraries.

To run compiled applications on Microsoft* Windows* OS, make sure that Microsoft* Visual C++ 2017 Redistributable and Intel® C++ Compiler 2017 Redistributable packages are installed and <INSTALL_DIR>/bin/intel64/Release/*.dll files are placed to the application folder or accessible via PATH% environment variable.