Merge branch 'openvinotoolkit:master' into master

Author: ieliz

data/dataset_definitions.yml

@@ -155,6 +155,16 @@ datasets:
       - name: AP
         type: coco_orig_keypoints_precision
 
+  - name: COCO2017_mask_rcnn_person
+    annotation_conversion:
+      converter: mscoco_mask_rcnn
+      annotation_file: instances_val2017_persons.json
+      has_background: True
+      sort_annotations: True
+    annotation: mscoco_mask_rcnn_person.pickle
+    dataset_meta: mscoco_mask_rcnn_person.json
+    data_source: val2017
+
   - name: imagenet_1000_classes
     annotation_conversion:
       converter: imagenet

demos/classification_demo/python/README.md

@@ -50,6 +50,8 @@ omz_converter --list models.lst
 * efficientnet-b5-pytorch
 * efficientnet-b7_auto_aug
 * efficientnet-b7-pytorch
+* efficientnet-v2-b0
+* efficientnet-v2-s
 * googlenet-v1
 * googlenet-v1-tf
 * googlenet-v2

demos/classification_demo/python/models.lst

@@ -18,6 +18,8 @@ efficientnet-b5
 efficientnet-b5-pytorch
 efficientnet-b7_auto_aug
 efficientnet-b7-pytorch
+efficientnet-v2-b0
+efficientnet-v2-s
 googlenet-v1
 googlenet-v1-tf
 googlenet-v2

demos/handwritten_text_recognition_demo/python/README.md

@@ -77,7 +77,10 @@ Options:
                         Optional. Top k steps in looking up the decoded
                         character, until a designated one is found
   -ob OUTPUT_BLOB, --output_blob OUTPUT_BLOB
-                        Optional. Name of the output layer of the model. Default is 'output'
+                        Optional. Name of the output layer of the model.
+                        Default is None, in which case the demo will read
+                        the output name from the model, assuming there is
+                        only 1 output layer
 ```
 
 The decoding char list files provided within Open Model Zoo and for Japanese it is the `<omz_dir>/data/dataset_classes/kondate_nakayosi.txt` file, while for Simplified Chinese it is the `<omz_dir>/data/dataset_classes/scut_ept.txt` file, and for English it is the `<omz_dir>/data/dataset_classes/gnhk.txt` file. For example, to do inference on a CPU with the OpenVINO&trade; toolkit pre-trained `handwritten-japanese-recognition-0001` model, run the following command:

demos/tests/run_tests.py

@@ -191,7 +191,12 @@ def main():
     model_info_list = json.loads(subprocess.check_output(
         [sys.executable, '--', str(auto_tools_dir / 'info_dumper.py'), '--all'],
         universal_newlines=True))
-    model_info = {model['name']: model for model in model_info_list}
+
+    model_info = {}
+    for model_data in model_info_list:
+        models_list = model_data['model_stages'] if model_data['model_stages'] else [model_data]
+        for model in models_list:
+            model_info[model['name']] = model
 
     if args.demos is not None:
         names_of_demos_to_test = set(args.demos.split(','))
@@ -219,11 +224,14 @@ def main():
             print()
             demo.set_precisions(args.precisions, model_info)
 
-            declared_model_names = {model['name']
-                for model in json.loads(subprocess.check_output(
+            declared_model_names = set()
+            for model_data in json.loads(subprocess.check_output(
                     [sys.executable, '--', str(auto_tools_dir / 'info_dumper.py'),
                         '--list', str(demo.models_lst_path(demos_dir))],
-                    universal_newlines=True))}
+                    universal_newlines=True)):
+                models_list = model_data['model_stages'] if model_data['model_stages'] else [model_data]
+                for model in models_list:
+                    declared_model_names.add(model['name'])
 
             with temp_dir_as_path() as temp_dir:
                 arg_context = ArgContext(

models/intel/action-recognition-0001/composite-model.yml

@@ -1 +1,6 @@
+description: >-
+  General-purpose action recognition model for Kinetics-400 dataset based on Video
+  Transformer Network approach.
 task_type: action_recognition
+framework: dldt
+license: https://raw.githubusercontent.com/openvinotoolkit/open_model_zoo/master/LICENSE

models/intel/device_support.md

@@ -53,12 +53,14 @@
 | instance-segmentation-security-0228 | YES |    |    |
 | instance-segmentation-security-1039 | YES |    |    |
 | instance-segmentation-security-1040 | YES |    |    |
+| instance-segmentation-person-0007 | YES |    |    |
 | landmarks-regression-retail-0009 | YES | YES | YES |
 | license-plate-recognition-barrier-0001 | YES | YES | YES |
 | machine-translation-nar-de-en-0002 | YES | YES |    |
 | machine-translation-nar-en-de-0002 | YES | YES |    |
 | machine-translation-nar-en-ru-0001 | YES | YES |    |
 | machine-translation-nar-ru-en-0001 | YES | YES |    |
+| noise-suppression-denseunet-ll-0001 | YES |    |    |
 | noise-suppression-poconetlike-0001 | YES |    |    |
 | pedestrian-and-vehicle-detector-adas-0001 | YES | YES | YES |
 | pedestrian-detection-adas-0002 | YES | YES | YES |

models/intel/driver-action-recognition-adas-0002/composite-model.yml

@@ -1 +1,5 @@
+description: >-
+  Video Transformer Network for driver action recognition.
 task_type: action_recognition
+framework: dldt
+license: https://raw.githubusercontent.com/openvinotoolkit/open_model_zoo/master/LICENSE

models/intel/formula-recognition-medium-scan-0001/composite-model.yml

@@ -1 +1,5 @@
+description: >-
+  Latex formula recognition model.
 task_type: token_recognition
+framework: dldt
+license: https://raw.githubusercontent.com/openvinotoolkit/open_model_zoo/master/LICENSE

models/intel/formula-recognition-polynomials-handwritten-0001/composite-model.yml

@@ -1 +1,5 @@
+description: >-
+  Latex formula recognition model.
 task_type: token_recognition
+framework: dldt
+license: https://raw.githubusercontent.com/openvinotoolkit/open_model_zoo/master/LICENSE