Linux Retail content with double MYRIAD fix

Video_Analytics/Python/Age_Gender_Detection.md

@@ -30,49 +30,46 @@ parser.add_argument("-d_ag", "--device_ag",
                     type=str)
 #TODO Head_Pose command line arguments
   ```
-### 2. Inializing Plugin for Myriad for Age amd Gender
+### 2. Initializing Plugin for Age and Gender
 Initialize Plugin for Myriad for Age and Gender since it cannot be loaded twice.
-
+- Replace **#TODO Age_Gender_Detection 1**
+- Paste the following lines
 ```python
-      MYRIAD_plugin_ag = IEPlugin(args.device_ag.upper(),args.plugin_dir)
-      #TODO Initializing Plugin for Myriad for Head Pose
+
+    if args.model and args.ag_model:
+        plugin_ag,ag_net = load_model("Age/Gender Recognition",args.ag_model,args.device_ag,args.plugin_dir,1,2,args.cpu_extension)
+        if args.device == args.device_ag:
+            plugin_ag = plugin
+            #TODO Head_Pose_Detection 2
+
 ```
 
 ### 3. Load Pre-trained Optimized Model for Age and Gender Inferencing
 
 In previous step, CPU is selected as plugin device. Now, load pre-trained optimized model for age and gender detection inferencing on CPU.
-- Replace **#TODO Age_Gender_Detection 1**
+- Replace **#TODO Age_Gender_Detection 2**
 - Paste the following lines
 
 ```python
+
+
     # age and gender   
     if args.model and args.ag_model:
 
        age_enabled =True
-       #log.info("Loading network files for Age/Gender Recognition")
-       plugin,ag_net=load_model("Age/Gender Recognition",args.ag_model,args.device_ag.upper(),args.plugin_dir,1,2,args.cpu_extension)
        age_input_blob=next(iter(ag_net.inputs))
        age_out_blob=next(iter(ag_net.outputs))
-
-
-       if ((args.device_ag.upper() == "MYRIAD") and (not args.device.upper() == "MYRIAD")):
-           age_exec_net = MYRIAD_plugin_ag.load(network=ag_net, num_requests=2)
-       elif (args.device_ag == "MYRIAD"):
-           age_exec_net = MYRIAD_plugin.load(network=ag_net, num_requests=2)
-       else :
-           age_exec_net = plugin.load(network=ag_net, num_requests=2)      
-
-
+       age_exec_net = plugin.load(network=ag_net, num_requests=2)      
        ag_n, ag_c, ag_h, ag_w = ag_net.inputs[input_blob].shape
        del ag_net
 
-   #TODO Head_Pose_Detection 2
+   #TODO Head_Pose_Detection 3
 
 ```
 
 ### 4. Initialize the parameters
 Here initialize the parameters which are required to process the output.
-- Replace **#TODO Age_Gender_Detection 2**
+- Replace **#TODO Age_Gender_Detection 3**
 - Paste the following lines
 
 ```python
@@ -87,7 +84,7 @@ attentivityindex = 0
 ### 5. Resetting the parameters for each frame
 The initialized parameters which are required to process the output are reset to zero.
 
-- Replace **#TODO Age_Gender_Detection 3**
+- Replace **#TODO Age_Gender_Detection 4**
 - Paste the following lines
 
 ```python
@@ -100,7 +97,7 @@ attentivityindex=0
 ### 6. Process Face detection Inference Results
 At this stage face detection Inference results will be available for further processing. Here, identified face will be clipped off and will be used for identifying age and gender in next request for inferencing.
 
-- Replace **#TODO Age_Gender_Detection 4**
+- Replace **#TODO Age_Gender_Detection 5**
 - Paste the following lines
 
 ```python
@@ -117,13 +114,13 @@ if age_enabled:
     if age_exec_net.requests[cur_request_id].wait(-1) == 0:
         age_inf_end = time.time()
         age_inf_time=age_inf_end - age_inf_start
-#TODO Head_Pose_Detection 3          
+#TODO Head_Pose_Detection 4          
 
 ```
 
 ### 7. Process Age and Gender detection Results for display
 Now we got result for Face, Age and Gender detection. We can customize the output and display this on the screen
-- Replace **#TODO Age_Gender_Detection 5**
+- Replace **#TODO Age_Gender_Detection 6**
 - Paste the following lines
 
 ```python

Video_Analytics/Python/Explore_OpenVINO.md

@@ -66,7 +66,7 @@ Upon the start-up, the demo application reads command line parameters and loads
 - **Case 1:** To run the sample application on **CPU**, execute the following commands:
 
 ```
-python3 object_detection_demo_ssd_async/object_detection_demo_ssd_async.py -i cam -m /opt/intel/openvino/deployment_tools/tools/model_downloader/Retail/object_detection/face/sqnet1.0modif-ssd/0004/dldt/face-detection-retail-0004.xml -l/opt/intel/openvino/deployment_tools/inference_engine/samples/build/intel64/Release/lib/libcpu_extension.so -d CPU
+python3 object_detection_demo_ssd_async/object_detection_demo_ssd_async.py -i cam -m /opt/intel/openvino/deployment_tools/tools/model_downloader/Retail/object_detection/face/sqnet1.0modif-ssd/0004/dldt/face-detection-retail-0004.xml -l /opt/intel/openvino/deployment_tools/inference_engine/samples/build/intel64/Release/lib/libcpu_extension.so -d CPU
 
 ```
 
@@ -76,7 +76,7 @@ Next, we repeat the exercise with **GPU** and observe the performance.
 - **case2:** To run the sample application on **GPU**, execute the following commands;
 
 ```
-python3 object_detection_demo_ssd_async/object_detection_demo_ssd_async.py -i cam -m /opt/intel/openvino/deployment_tools/tools/model_downloader/Retail/object_detection/face/sqnet1.0modif-ssd/0004/dldt/face-detection-retail-0004.xml -l/opt/intel/openvino/deployment_tools/inference_engine/samples/build/intel64/Release/lib/libcpu_extension.so -d GPU
+python3 object_detection_demo_ssd_async/object_detection_demo_ssd_async.py -i cam -m /opt/intel/openvino/deployment_tools/tools/model_downloader/Retail/object_detection/face/sqnet1.0modif-ssd/0004/dldt/face-detection-retail-0004.xml -l /opt/intel/openvino/deployment_tools/inference_engine/samples/build/intel64/Release/lib/libcpu_extension.so -d GPU
 
 ```
 - **Output and Performance**

Video_Analytics/Python/Face_detection.md

@@ -139,33 +139,26 @@ First, the pre-trained model XML files have been optimized and generated by the
 
 ```python
 
+
     log.basicConfig(format="[ %(levelname)s ] %(message)s", level=log.INFO, stream=sys.stdout)
     args = build_argparser().parse_args()
     age_enabled = False
     headPose_enabled = False
     #TODO Cloud_Integration 2
 
-
-    MYRIAD_plugin = IEPlugin(args.device.upper(),args.plugin_dir)
-    #TODO Initializing Plugin for Myriad for Age and Gender
+    #Make sure only one IEPlugin was created for one type of device
+    plugin,net = load_model("Face Detection",args.model,args.device,args.plugin_dir,1,1,args.cpu_extension)    
+    #TODO Age_Gender_Detection 1
 
 
     log.info("Reading IR...")
     # Face detection
-    #log.info("Loading network files for Face Detection")
-
-    plugin,net=load_model("Face Detection",args.model,args.device.upper(),args.plugin_dir,1,1,args.cpu_extension)
     input_blob = next(iter(net.inputs))
     out_blob = next(iter(net.outputs))
-
-    if (args.device.upper() == "MYRIAD"):
-        exec_net = MYRIAD_plugin.load(network=net, num_requests=2)
-    else :
-        exec_net = plugin.load(network=net, num_requests=2)
-
+    exec_net = plugin.load(network=net, num_requests=2)
     n, c, h, w = net.inputs[input_blob].shape
     del net
-    #TODO Age_Gender_Detection 1
+    #TODO Age_Gender_Detection 2
 
   ```
 
@@ -197,7 +190,7 @@ log.info("Starting inference ...")
 log.info("To stop the demo execution press Esc button")
 is_async_mode = True
 render_time = 0
-#TODO Age_Gender_Detection 2
+#TODO Age_Gender_Detection 3
 decode_time = 0
 visual_time = 0
 framesCounter = 0
@@ -215,7 +208,7 @@ decode_prev_time = decode_prev_finish - decode_prev_start
 
 ```python
 while cap.isOpened():
-    #TODO Age_Gender_Detection 3
+    #TODO Age_Gender_Detection 4
     analytics_time = 0
     decode_next_start = time.time()
     ret, frame = cap.read()
@@ -258,16 +251,16 @@ while cap.isOpened():
                 height = ymax - ymin
                 width = xmax -xmin  
 
-                #TODO Age_Gender_Detection 4
+                #TODO Age_Gender_Detection 5
                 visual_start = time.time()                                   
                 if args.no_show==False:    
-                    #TODO Age_Gender_Detection 5
+                    #TODO Age_Gender_Detection 6
                     class_id = int(obj[1])                                     
                     # Draw box and label\class_id
                     color = (min(class_id * 12.5, 255), min(class_id * 7, 255), min(class_id * 5, 255))
                     cv2.rectangle(frame, (xmin, ymin), (xmax, ymax), (255,10,10), 2)
                     det_label = labels_map[class_id] if labels_map else str(class_id)
-                    #TODO Head_Pose_Detection 4
+                    #TODO Head_Pose_Detection 5
 
                     render_time_message = "OpenCV cap/rendering time: {:.2f} ms".format(render_time * 1000)
                     inf_time_message = "Face Detection time: {:.2f} ms ({:.2f} fps)".format((det_time * 1000),1/(det_time))           

Video_Analytics/Python/Head_Pose_Detection.md

@@ -109,52 +109,54 @@ def drawAxes(pitch,yaw,roll,cpoint,frame):
 ```
 ### 3. Initializing Plugin for Myriad for Head Pose
 Initialize Plugin for Myriad for Head Pose
+- Replace **#TODO Head_Pose_Detection 2**
+- Paste the following lines
 
 ```python
-    MYRIAD_plugin_hp = IEPlugin(args.device_hp.upper(),args.plugin_dir)
+
+            if args.model and args.hp_model:
+                plugin_hp,hp_net=load_model("Head Pose Estimation",args.hp_model,args.device_hp,args.plugin_dir,1,3,args.cpu_extension)
+                if args.device == args.device_hp:                    
+                    plugin_hp = plugin
+
+        else :
+            if args.model and args.hp_model:
+                plugin_hp,hp_net=load_model("Head Pose Estimation",args.hp_model,args.device_hp,args.plugin_dir,1,3,args.cpu_extension)
+                if args.device_ag == args.device_hp:
+                    plugin_hp = plugin_ag
+                elif args.device == args.device_hp:
+                    plugin_hp = plugin         
+
+    else :
+        if args.model and args.hp_model:
+            plugin_hp,hp_net=load_model("Head Pose Estimation",args.hp_model,args.device_hp,args.plugin_dir,1,3,args.cpu_extension)    
+            if args.device == args.device_hp:                    
+                plugin_hp = plugin                        
+
 ```
 
 ### 4. Load Pre-trained Optimized Model for HeadPose Inferencing
 Load pre-retained optimized model for head pose detection on CPU as CPU is already selected as plugin device previously.
-- Replace **#TODO Head_Pose_Detection 2**
+- Replace **#TODO Head_Pose_Detection 3**
 - Paste the following lines
 
 ```python
 
     # Head Pose  
     if args.model and args.hp_model:
         headPose_enabled = True
-        #log.info("Loading network files for Head Pose Estimation")
-        plugin,hp_net=load_model("Head Pose Estimation",args.hp_model,args.device_hp.upper(),args.plugin_dir,1,3,args.cpu_extension)
         hp_input_blob=next(iter(hp_net.inputs))
         hp_out_blob=next(iter(hp_net.outputs))
-
-        if (args.device_hp.upper() == "MYRIAD" and not args.device.upper() =="MYRIAD" and not args.device_ag.upper() == "MYRIAD"):
-            hp_exec_net = MYRIAD_plugin_hp.load(network=hp_net, num_requests=2)
-
-        elif (args.device_hp.upper() == "MYRIAD"):
-            if (args.device_ag.upper() == "MYRIAD"):
-                if (args.device.upper() == "MYRIAD"):
-                    hp_exec_net = MYRIAD_plugin.load(network=hp_net, num_requests=2)
-                else :
-                    hp_exec_net = MYRIAD_plugin_ag.load(network=hp_net, num_requests=2)
-            elif (args.device.upper() == "MYRIAD"):
-                hp_exec_net = MYRIAD_plugin.load(network=hp_net, num_requests=2)
-            else :
-                hp_exec_net = MYRIAD_plugin_hp.load(network=hp_net, num_requests=2)
-
-        else :
-            hp_exec_net = plugin.load(network=hp_net, num_requests=2)
-
-
+        hp_exec_net=plugin_hp.load(network=hp_net, num_requests=2)
         hp_n, hp_c, hp_h, hp_w = hp_net.inputs[input_blob].shape
         del hp_net
 
+
 ```
 
 ### 5. Use Identified Face for HeadPose Detection
 Clip the identified Faces and send inference request for detecting head pose
-- Replace **#TODO Head_Pose_Detection 3**
+- Replace **#TODO Head_Pose_Detection 4**
 - Paste the following lines
 
 ```python
@@ -178,7 +180,7 @@ analytics_time = age_inf_time + hp_inf_time
 
 ### 6. Calculate attentivity index
 Now, the attentivityindex will be calculated using Yaw angle available from head pose inference result.
-- Replace **#TODO Head_Pose_Detection 4**
+- Replace **#TODO Head_Pose_Detection 5**
 - Paste the following lines
 
 ```python