- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
The current version support MASKRCNN running on CPU. I am trying to run a mobile-net backbone based mask RCNN. the original source is @https://github.com/GustavZ/Mobile_Mask_RCNN , with Keras and tensor flow model, I use cmd
python3 mo_tf.py --input_model model.pb , then it reports "Graph contains a cycle"
I change the command to
python3 mo_tf.py --input_model model.pb --tensorflow_use_custom_operations_config extensions/front/tf/mask_rcnn_support.json
it reports "Generated IR model" , yet the .bin is empty .
Any suggestion on it?
If you need the pb model. I could send it out.
thanks,
Ling
Link Copied
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Dear Liu,
you are missing one argument in your command line --tensorflow_object_detection_api_pipeline_config, please look at our documentation on how we support Mask RCNN: computer_vision_sdk_2018.3.343/deployment_tools/documentation/docs/TensorFlowObjectDetectionAPIModels.html for more details
Also, if the model is not from the model zoo, you might want to try with the mask_rcnn_support_api_v1.7.json
file.
Best,
Severine
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Dear Habert:
thanks for your quick response. I add the argument --tensorflow_object_detection_api_pipeline_config , and the model could be generated, yet the bin file is empty. I use the tool to open my pb model and the official pb model from model zoo, they two looks quite different. It seems the official pb model has more detail layers and lots of customer defined layers. Could you tell me where is the source of the official pb model? or how to generate the pb model like yours?
thanks,
Ling
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Dear Habert:
I read the document computer_vision_sdk_2018.3.343/deployment_tools/documentation/docs/TensorFlowObjectDetectionAPIModels as you suggested. If I understand correctly, now we only support Tensorflow opject detection API models from the object detection model zoo. Currently the MASK-RCNN is reset101/50/inception backboned. Our goal is to implement a mobilenet backbone MASK RCNN, my question is:
If I use tensorflow object detection API to build the mobilenet backbone mask RCNN, does model optimizer could transfer it correctly and we could run it on OPENVINO?
Since the ssd-mobilenet-* network could be supported now, I guess Mask_RCNN_mobilenet* should be worked, right? DO we have any plan to support it?
thanks,
Ling
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page