[PP] Addded ability to preprocess inputs into plugin (#857)

desired format

changed InferRequestInternal:
 - added _deviceInputs member to store plugin desired perprocessing
   targets
 - added default argument to preProcessingRequired to describe plugin
   specific desired preprocessing target
 - SetBlob and GetBlob to deal with plugin desired preprocessing targets
   (_deviceInputs)
 - added addInputPreProcessingFor helper method to avoid code
   duplication

changed TEMPLATE plugin to use new functionality:
 - removed explicit presicion conversion (to use built-in one of
   InferRequestInternal)
 - _networkInputBlobs to use InferRequestInternal::_deviceInputs

docs/template_plugin/src/template_infer_request.cpp

@@ -112,7 +112,7 @@ static void AllocateImpl(const BlobDataMap& blobDataMap,
 
 void TemplateInferRequest::allocateBlobs() {
     auto&& parameters = _executableNetwork->_function->get_parameters();
-    AllocateImpl(_networkInputs, _inputs, _networkInputBlobs, [&] (const std::string& blobName) {
+    AllocateImpl(_networkInputs, _inputs, _deviceInputs, [&] (const std::string& blobName) {
         return parameters.at(_executableNetwork->_inputIndex.at(blobName))->get_element_type();
     });
     auto&& results = _executableNetwork->_function->get_results();
@@ -176,21 +176,14 @@ void TemplateInferRequest::inferPreprocess() {
     auto start = Time::now();
     // NOTE: After InferRequestInternal::execDataPreprocessing call
     //       input can points to other memory region than it was allocated in constructor.
-    InferRequestInternal::execDataPreprocessing(_inputs);
-    for (auto&& input : _inputs) {
-        auto inputBlob = input.second;
-        auto networkInput = _networkInputBlobs[input.first];
-        if (inputBlob->getTensorDesc().getPrecision() == networkInput->getTensorDesc().getPrecision()) {
-            networkInput = inputBlob;
-        } else {
-            blobCopy(inputBlob, networkInput);
-        }
-        auto index = _executableNetwork->_inputIndex[input.first];
+    InferRequestInternal::execDataPreprocessing(_deviceInputs);
+    for (auto&& networkInput : _deviceInputs) {
+        auto index = _executableNetwork->_inputIndex[networkInput.first];
         const auto& parameter = _parameters[index];
         const auto& parameterShape = parameter->get_shape();
         const auto& parameterType = parameter->get_element_type();
         _inputTensors[index] = _executableNetwork->_plugin->_backend->create_tensor(parameterType, parameterShape,
-            InferenceEngine::as<InferenceEngine::MemoryBlob>(networkInput)->rmap().as<void*>());
+            InferenceEngine::as<InferenceEngine::MemoryBlob>(networkInput.second)->rmap().as<void*>());
     }
     for (auto&& output : _outputs) {
         auto outputBlob = output.second;

docs/template_plugin/src/template_infer_request.hpp

@@ -63,7 +63,6 @@ private:
     // for performance counters
     std::array<std::chrono::duration<float, std::micro>, numOfStages>   _durations;
 
-    InferenceEngine::BlobMap                                _networkInputBlobs;
     InferenceEngine::BlobMap                                _networkOutputBlobs;
     ngraph::ParameterVector                                 _parameters;
     ngraph::ResultVector                                    _results;

docs/template_plugin/tests/functional/shared_tests_instances/behavior/preprocessing.cpp

@@ -22,7 +22,7 @@ const std::vector<std::map<std::string, std::string>> configs = {
 INSTANTIATE_TEST_CASE_P(PreprocessingPrecisionConvertTestsViaSetInput, PreprocessingPrecisionConvertTest,
                         ::testing::Combine(
                                 ::testing::ValuesIn(inputPrecisions),
-                                ::testing::Values(1, 2, 3, 4, 5),   // Number of input tensor channels
+                                ::testing::Values(4),   // Number of input tensor channels
                                 ::testing::Values(true),            // Use SetInput
                                 ::testing::Values(CommonTestUtils::DEVICE_TEMPLATE),
                                 ::testing::ValuesIn(configs)),
@@ -31,7 +31,7 @@ INSTANTIATE_TEST_CASE_P(PreprocessingPrecisionConvertTestsViaSetInput, Preproces
 INSTANTIATE_TEST_CASE_P(PreprocessingPrecisionConvertTestsViaGetBlob, PreprocessingPrecisionConvertTest,
                         ::testing::Combine(
                                 ::testing::ValuesIn(inputPrecisions),
-                                ::testing::Values(4, 5),       // Number of input tensor channels (blob_copy only supports 4d and 5d tensors)
+                                ::testing::Values(4),       // Number of input tensor channels (blob_copy only supports 4d and 5d tensors)
                                 ::testing::Values(false),      // use GetBlob
                                 ::testing::Values(CommonTestUtils::DEVICE_TEMPLATE),
                                 ::testing::ValuesIn(configs)),