OpenVino 2020 (#1269)

* added support for OpenVINO 2020

* fixed dextr and tf_annotation

Co-authored-by: Andrey Zhavoronkov <andrey.zhavoronkov@intel.com>

CHANGELOG.md

@@ -52,6 +52,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - React & Redux & Antd based dashboard
 - Yolov3 interpretation script fix and changes to mapping.json
 - YOLO format support ([#1151](https://github.com/opencv/cvat/pull/1151))
+- Added support for OpenVINO 2020
 
 ### Fixed
 - Exception in Git plugin [#826](https://github.com/opencv/cvat/issues/826)

cvat/apps/auto_annotation/inference_engine.py

@@ -2,7 +2,7 @@
 #
 # SPDX-License-Identifier: MIT
 
-from openvino.inference_engine import IENetwork, IEPlugin
+from openvino.inference_engine import IENetwork, IEPlugin, IECore, get_version
 
 import subprocess
 import os
@@ -19,7 +19,20 @@ def _check_instruction(instruction):
     )
 
 
-def make_plugin():
+def make_plugin_or_core():
+    version = get_version()
+    use_core_openvino = False
+    try:
+        major, minor, reference = [int(x) for x in version.split('.')]
+        if major >= 2 and minor >= 1 and reference >= 37988:
+            use_core_openvino = True
+    except Exception:
+        pass
+
+    if use_core_openvino:
+        ie = IECore()
+        return ie
+
     if _IE_PLUGINS_PATH is None:
         raise OSError('Inference engine plugin path env not found in the system.')
 

cvat/apps/auto_annotation/model_loader.py

@@ -8,25 +8,22 @@ import cv2
 import os
 import numpy as np
 
-from cvat.apps.auto_annotation.inference_engine import make_plugin, make_network
+from cvat.apps.auto_annotation.inference_engine import make_plugin_or_core, make_network
 
 class ModelLoader():
     def __init__(self, model, weights):
         self._model = model
         self._weights = weights
 
-        IE_PLUGINS_PATH = os.getenv("IE_PLUGINS_PATH")
-        if not IE_PLUGINS_PATH:
-            raise OSError("Inference engine plugin path env not found in the system.")
-
-        plugin = make_plugin()
+        core_or_plugin = make_plugin_or_core()
         network = make_network(self._model, self._weights)
 
-        supported_layers = plugin.get_supported_layers(network)
-        not_supported_layers = [l for l in network.layers.keys() if l not in supported_layers]
-        if len(not_supported_layers) != 0:
-            raise Exception("Following layers are not supported by the plugin for specified device {}:\n {}".
-                      format(plugin.device, ", ".join(not_supported_layers)))
+        if getattr(core_or_plugin, 'get_supported_layers', False):
+            supported_layers = core_or_plugin.get_supported_layers(network)
+            not_supported_layers = [l for l in network.layers.keys() if l not in supported_layers]
+            if len(not_supported_layers) != 0:
+                raise Exception("Following layers are not supported by the plugin for specified device {}:\n {}".
+                          format(core_or_plugin.device, ", ".join(not_supported_layers)))
 
         iter_inputs = iter(network.inputs)
         self._input_blob_name = next(iter_inputs)
@@ -45,7 +42,12 @@ class ModelLoader():
         if self._input_blob_name in info_names:
             self._input_blob_name = next(iter_inputs)
 
-        self._net = plugin.load(network=network, num_requests=2)
+        if getattr(core_or_plugin, 'load_network', False):
+            self._net = core_or_plugin.load_network(network,
+                                                    "CPU",
+                                                    num_requests=2)
+        else:
+            self._net = core_or_plugin.load(network=network, num_requests=2)
         input_type = network.inputs[self._input_blob_name]
         self._input_layout = input_type if isinstance(input_type, list) else input_type.shape
 

cvat/apps/dextr_segmentation/dextr.py

@@ -3,7 +3,7 @@
 #
 # SPDX-License-Identifier: MIT
 
-from cvat.apps.auto_annotation.inference_engine import make_plugin, make_network
+from cvat.apps.auto_annotation.inference_engine import make_plugin_or_core, make_network
 
 import os
 import cv2
@@ -32,12 +32,15 @@ class DEXTR_HANDLER:
     def handle(self, im_path, points):
         # Lazy initialization
         if not self._plugin:
-            self._plugin = make_plugin()
+            self._plugin = make_plugin_or_core()
             self._network = make_network(os.path.join(_DEXTR_MODEL_DIR, 'dextr.xml'),
                 os.path.join(_DEXTR_MODEL_DIR, 'dextr.bin'))
             self._input_blob = next(iter(self._network.inputs))
             self._output_blob = next(iter(self._network.outputs))
-            self._exec_network = self._plugin.load(network=self._network)
+            if getattr(self._plugin, 'load_network', False):
+                self._exec_network = self._plugin.load_network(self._network, 'CPU')
+            else:
+                self._exec_network = self._plugin.load(network=self._network)
 
         image = PIL.Image.open(im_path)
         numpy_image = np.array(image)

cvat/apps/tf_annotation/views.py

@@ -30,7 +30,7 @@ def load_image_into_numpy(image):
 
 
 def run_inference_engine_annotation(image_list, labels_mapping, treshold):
-    from cvat.apps.auto_annotation.inference_engine import make_plugin, make_network
+    from cvat.apps.auto_annotation.inference_engine import make_plugin_or_core, make_network
 
     def _normalize_box(box, w, h, dw, dh):
         xmin = min(int(box[0] * dw * w), w)
@@ -44,11 +44,14 @@ def run_inference_engine_annotation(image_list, labels_mapping, treshold):
     if MODEL_PATH is None:
         raise OSError('Model path env not found in the system.')
 
-    plugin = make_plugin()
+    core_or_plugin = make_plugin_or_core()
     network = make_network('{}.xml'.format(MODEL_PATH), '{}.bin'.format(MODEL_PATH))
     input_blob_name = next(iter(network.inputs))
     output_blob_name = next(iter(network.outputs))
-    executable_network = plugin.load(network=network)
+    if getattr(core_or_plugin, 'load_network', False):
+        executable_network = core_or_plugin.load_network(network, 'CPU')
+    else:
+        executable_network = core_or_plugin.load(network=network)
     job = rq.get_current_job()
 
     del network