model-commands.md

Model commands: tools, model references and run commands

additional references:

• convert main page: https://docs.openvinotoolkit.org/latest/openvino_docs_MO_DG_prepare_model_convert_model_Converting_Model.html
• detailed tf convert page: https://docs.openvinotoolkit.org/latest/openvino_docs_MO_DG_prepare_model_convert_model_tf_specific_Convert_Object_Detection_API_Models.html tensorflow models from intel page: https://docs.openvinotoolkit.org/latest/openvino_docs_MO_DG_prepare_model_convert_model_Convert_Model_From_TensorFlow.html#supported_topologies
• caffee model names: https://docs.openvinotoolkit.org/latest/openvino_docs_MO_DG_prepare_model_convert_model_Convert_Model_From_Caffe.html
• model downloader hint: https://community.intel.com/t5/Intel-Distribution-of-OpenVINO/Build-and-convert-Caffe-SSD300-VGG16-model-to-IR/td-p/1165772

script tools used:

• wget for downloads

  wget
  

• tar for unpacking of *.tar.gz files

  tar -xvf
  

models used:

• model 01:

  link: http://download.tensorflow.org/models/object_detection/ssd_mobilenet_v2_coco_2018_03_29.tar.gz

  /opt/intel/openvino/deployment_tools/model_optimizer/mo_tf.py --input_model=ssd_mobilenet_v2_coco_2018_03_29/frozen_inference_graph.pb --output=detection_classes,detection_scores,detection_boxes,num_detections --tensorflow_object_detection_api_pipeline_config=ssd_mobilenet_v2_coco_2018_03_29/pipeline.config --tensorflow_use_custom_operations_config=/opt/intel/openvino/deployment_tools/model_optimizer/extensions/front/tf/ssd_v2_support.json --input=image_tensor --input_shape=[1,300,300,3] --reverse_input_channels --data_type FP16
  

  python /opt/intel/openvino/deployment_tools/model_optimizer/mo.py --input_model frozen_inference_graph.pb --tensorflow_object_detection_api_pipeline_config pipeline.config --reverse_input_channels --tensorflow_use_custom_operations_config /opt/intel/openvino/deployment_tools/model_optimizer/extensions/front/tf/ssd_v2_support.json
  

  result: **SUCCESS** [ SUCCESS ] Total execution time: 71.86 seconds. 
  

• model 02:

  link; http://download.tensorflow.org/models/object_detection/ssdlite_mobilenet_v2_coco_2018_05_09.tar.gz

  /opt/intel/openvino/deployment_tools/model_optimizer/mo_tf.py --input_model=ssdlite_mobilenet_v2_coco_2018_05_09/frozen_inference_graph.pb --output=detection_classes,detection_scores,detection_boxes,num_detections --tensorflow_object_detection_api_pipeline_config=ssdlite_mobilenet_v2_coco_2018_05_09/pipeline.config -tensorflow_use_custom_operations_config=/opt/intel/openvino/deployment_tools/model_optimizer/extensions/front/tf/ssd_v2_support.json --input=image_tensor --input_shape=[1,300,300,3] --reverse_input_channels --data_type FP16
  

  [ SUCCESS ] Generated IR model.
  [ SUCCESS ] XML file: /home/workspace/./frozen_inference_graph.xml
  [ SUCCESS ] BIN file: /home/workspace/./frozen_inference_graph.bin
  [ SUCCESS ] Total execution time: 52.28 seconds. 
  

• model 03:

  link: found over model downloader

  go to downloader:
  cd /opt/intel/openvino_2019.3.376/deployment_tools/tools/model_downloader/
  list all available models:
  python downloader.py --print_all
  download specific:
  sudo python downloader.py --name ssd300 -o /home/workspace/

  python /opt/intel/openvino/deployment_tools/model_optimizer/mo.py --input_model SSD_300x300_ft/VGG_VOC0712Plus_SSD_300x300_ft_iter_160000.caffemodel --input_proto SSD_300x300_ft/deploy.prototxt
  

• model 04:

  link: http://download.tensorflow.org/models/object_detection/faster_rcnn_inception_v2_coco_2018_01_28.tar.gz

  python /opt/intel/openvino/deployment_tools/model_optimizer/mo.py --input_model faster_rcnn_inception_v2_coco_2018_01_28/frozen_inference_graph.pb --tensorflow_object_detection_api_pipeline_config faster_rcnn_inception_v2_coco_2018_01_28/pipeline.config --reverse_input_channels --tensorflow_use_custom_operations_config /opt/intel/openvino/deployment_tools/model_optimizer/extensions/front/tf/faster_rcnn_support.json --input_shape=[1,300,300,3]
  

  [ SUCCESS ] Generated IR model.
  [ SUCCESS ] XML file: /home/workspace/./frozen_inference_graph.xml
  [ SUCCESS ] BIN file: /home/workspace/./frozen_inference_graph.bin
  [ SUCCESS ] Total execution time: 148.33 seconds.
  

RUN!

• FRNN

  python main.py -m models/frozen_inference_graph_FRNN_IC_V2.xml -i resources/Pedestrian_Detect_2_1_1.mp4 -l /opt/intel/openvino/deployment_tools/inference_engine/lib/intel64/libcpu_extension_sse4.so -d CPU -pt 0.6 | ffmpeg -v warning -f rawvideo -pixel_format bgr24 -video_size 768x432 -framerate 24 -i - http://0.0.0.0:3004/fac.ffm
  
  
  

• SSDLite mobilenet

  python main.py -m models/frozen_inference_graph_SSDlite_MN_V2.xml -i resources/Pedestrian_Detect_2_1_1.mp4 -l /opt/intel/openvino/deployment_tools/inference_engine/lib/intel64/libcpu_extension_sse4.so -d CPU -pt 0.6 | ffmpeg -v warning -f rawvideo -pixel_format bgr24 -video_size 768x432 -framerate 24 -i - http://0.0.0.0:3004/fac.ffm
  

• VGG

  python main.py -m models/VGG_VOC0712Plus_SSD_300x300_ft_iter_160000.xml -i resources/Pedestrian_Detect_2_1_1.mp4 -l /opt/intel/openvino/deployment_tools/inference_engine/lib/intel64/libcpu_extension_sse4.so -pt 0.8 | ffmpeg -v warning -f rawvideo -pixel_format bgr24 -video_size 768x432 -framerate 24 -i - http://0.0.0.0:3004/fac.ffm
  

• SSD mobilenet

  python main.py -m models/frozen_inference_graph_SSD_MN_V2.xml -i resources/Pedestrian_Detect_2_1_1.mp4 -l /opt/intel/openvino/deployment_tools/inference_engine/lib/intel64/libcpu_extension_sse4.so -d CPU -pt 0.6 | ffmpeg -v warning -f rawvideo -pixel_format bgr24 -video_size 768x432 -framerate 24 -i - http://0.0.0.0:3004/fac.ffm

• INTEL model

  python main.py -m models/person-detection-retail-0013-FP32.xml -i resources/Pedestrian_Detect_2_1_1.mp4 -l /opt/intel/openvino/deployment_tools/inference_engine/lib/intel64/libcpu_extension_sse4.so -pt 0.6 | ffmpeg -v warning -f rawvideo -pixel_format bgr24 -video_size 768x432 -framerate 24 -i - http://0.0.0.0:3004/fac.ffm