Intel® Distribution of OpenVINO™ Toolkit
Community assistance about the Intel® Distribution of OpenVINO™ toolkit, OpenCV, and all aspects of computer vision-related on Intel® platforms.

Openvino accelerates compressed yolo model fails.

Feng__Zhang
Beginner
349 Views

Hi everybody,

I try to run yolov3 and tiny with openvino ,it is runing very well.

 

Then,I try to running a slim model.

The slim model was  modified filters.

(e.g. filters=256  to filters=128)

I successfully converted the model to IR.

I can running it ,but it hava not any result.

Then, I use "-t 0.001" to run "object_detection_demo_yolov3_async.exe" with my slim model,

I found that the confidence of the recognition results is extremely low, like 0.002, 0.003.

So this is also why before did not get the recognition result.

What should I do?

Thanks everybody.

 

----------------------------------------

My computer info:

Win 10 x64

Openvino R2 & R3 (all tried)

The attachment contains the modified yolov3 implementation of cfg and convert_weights_pb.py

 

 

 

 

0 Kudos
1 Reply
Feng__Zhang
Beginner
349 Views
Does anyone know?
0 Kudos
Reply