- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi,
I am getting following error while running object_detection_sample in the inference engine samples:
deepak@camry:~/test/intel/deep_learning_sdk_2017.1.0.2778/deployment_tools/inference_engine/bin/intel64/Release$ ./object_detection_sample -i ~/Desktop/image.bmp -m ~/test1/test/test-tiny.xml -d CPU
InferenceEngine:
API version ............ 1.0
Build .................. 2778
API version ............ 1.0
Build .................. 2017.0.2
Description ....... MKLDnnPlugin
Layer proposal not found in network
May I know the reason for this ?
Thank you.
Link Copied
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
This error means there is no layer called "proposal" in your model.
object_detection_sample can only process Faster-RCNN-type networks (that have a "proposal' layer inside). If you have some custom one, it won't work out-of-the-box because different object-detectors use different output data formats (there is no single standard).
What kind of model are you processing? Could you share it?
Thank you for your interest!
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi Ilya,
Ok. That's correct, my prototxt does not have any "proposal" layer.
I am using yolo-tiny-protxt. In this case what I need to do get the output ?
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi Deepak,
We are working on YOLO and YOLO-tiny support. These nets will be supported in the upcoming DL Toolkit release (that will be published this or next week).
Please, download the new DL toolkit release and let me know - I will help you with scoring YOLO-tiny.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Thank you Ilya. From where I can get the update on the new release ?
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi Deepak,
The Deep Learning Deployment Toolkit Beta 2017R2 has just released. You are welcome to download it and try. There you will find YOLO support and a sample.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I have installed new IntelSdkR2, Now running my code as follows, and getting the same (intel_cv_sdk_ubuntu_r2_2017.0.113):
deepak@Desktop:/opt/intel/computer_vision_sdk_2017.0.113/inference_engine/samples/object_detection_sample$ ./a.out -i ~/Downloads/num.jpg -m ~/testYolo/tiny/tiny.xml -d CPU
InferenceEngine:
API version ............ 1.0
Build .................. 4463
API version ............ 1.0
Build .................. lnx_2018.0.20170425
Description ....... MKLDnnPlugin
Layer proposal not found in network
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Please, use object_detection_sample_yolo, not object_detection_sample.

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page