Inference on YOLO

[R__Divyashree]

I was able to convert the .pb file of YOLO object detection API to IR files(.xml and .bin). But when I try to run inference using object_detection samples I am getting the following errors:

1)Using object_detection_sample_ssd:

./object_detection_sample_ssd -d "CPU" -m /home/gw2/IRfiles/tiny-yolo-voc-1c.xml -i /home/gw2/IRfiles/ATD_test_5.jpg

[ ERROR ] Can't find a DetectionOutput layer in the topology

2)Using object_detection_demo_ssd_async

./object_detection_demo_ssd_async -d "CPU" -m /home/gw2/IRfiles/tiny-yolo-voc-1c.xml -i /home/gw2/final.mp4

[ ERROR ] No such parameter name 'num_classes' for layer output/YoloRegion

3)Even object_detection_sample will not work

===\

Can someone guide on performing inference for YOLO?

[Shubha_R_Intel]

Divyashree, first, how did you get the YOLO *.pb file ? Was it based on yolov3 or yolov2 ?  Did you use deployment_tools\documentation\docs\YOLOTF.html as a guide ? The reason I am asking is because there is a known (not related to Intel)  bug that prevents one from converting a a Yolov3 to a tensor flow graph. More details about this bug here. I faced this bug  myself when I used darkflow as YOLOTF.html describes.

https://github.com/thtrieu/darkflow/issues/665

Thank you for using OpenVino and I shall await your reply.

Shubha

[r__divyashree1]

Hi Shubha, Thanks for your reply..

I used Tiny Yolo  and then I converted the corresponding cfg and weights to .pb.

Yes, I used YOLOTF.html as a reference to generate IR files.

Please, help me with the inference part.

[r__divyashree1]

Hello, I used Tiny YOLO of V2. Yes, I followed the YOLOTF.html as reference to generate the IR files.

Please, help me with the Inference Part.