I found an issue when I was doing inference using R5 version of OpenVINO toolkit on Intel Movidius Myriad X VPU. I used the following command to generate IR for my model:
Python mo.py --input_model E:\frcnn_resnet101\frozen_inference_graph.pb --tensorflow_use_custom_operations_config C:\Intel\computer_vision_sdk_2018.5.445\deployment_tools\model_optimizer\extensions\front\tf\faster_rcnn_support_api_v1.7.json --tensorflow_object_detection_api_pipeline_config E:\models-resnet\pipeline.config --data_type FP16
After getting the IR, when doing inference on Intel Movidius Myriad X VPU, the program stuck on the following step:
[ INFO ] Loading model to the plugin
More details are showing in the attachment.
Could anyone help to figure out what causes my problem?
For more complete information about compiler optimizations, see our Optimization Notice.