Summary
Steps to run "mask_rcnn_demo.exe"
Description
- Referred to Convert ONNX* Mask R-CNN Model to the Intermediate Representation from OpenVINO™ Documentation.
- Downloaded and converted mask_rcnn_R_50_FPN_1x model into IR
- Executed the command: mask_rcnn_demo.exe -i D:/hqx/yolact/test_Color.jpg -m D:/hqx/mask_rcnn_R_50_FPN_1x.xml -detection_output_name=DetectionOutput
- Received error message: [ ERROR ] Cannot find blob with name: DetectionOutput
- Executed different command: mask_rcnn_demo.exe -i D:/hqx/yolact/test_Color.jpg -m D:/hqx/mask_rcnn_R_50_FPN_1x.xml -detection_output_name=6849/sink_port_0
- Received error message: [ ERROR ] Cannot add output! Layer 6849/sink_port_0 wasn't found!
Resolution
- Go to mask_rcnn_demo repository which is located at: <INSTALL_DIR>\deployment_tools\open_model_zoo\demos\mask_rcnn_demo
- Open models.lst file
- Download suggested model
- mask_rcnn_inception_resnet_v2_atrous_coco
- mask_rcnn_inception_v2_coco
- mask_rcnn_resnet101_atrous_coco
- mask_rcnn_resnet50_atrous_coco
- Convert downloaded model with Model Optimizer
- Run mask_rcnn_demo.exe with the following command mask_rcnn_demo.exe -m "<PATH_TO_MODEL>\<MODEL>.xml" -i "<PATH_TO_IMAGE>\<IMAGE>.jpg"
Additional information
Refer TensorFlow* Object Detection Mask R-CNNs Segmentation C++ Demo for more information.