search

PINTO0309 / OpenVINO-YoloV3

YoloV3/tiny-YoloV3+RaspberryPi3/Ubuntu LaptopPC+NCS/NCS2+USB Camera+Python+OpenVINO
https://qiita.com/PINTO
Apache License 2.0
538 stars 167 forks source link

Decoding format of results stored in output blobs #67

Open pra-dan opened 4 years ago

pra-dan commented 4 years ago

Hi, I am working on the sane project, but locally and not on RPi or NCS. I am using YOLOv3-tiny and found that the output is 2 layers with

#Layer                                              |  Feature map shape
#detector/yolo-v3-tiny/Conv_12/BiasAdd/YoloRegion   |   (1, 255, 26, 26)
#detector/yolo-v3-tiny/Conv_9/BiasAdd/YoloRegion    |   (1, 255, 13, 13)

these results are flattened and then parsed using the below function: https://github.com/PINTO0309/OpenVINO-YoloV3/blob/b5caa1c2117749b84719c0cd9750c10c452ef471/openvino_tiny-yolov3_MultiStick_test.py#L54

Can you explain how these results are stored and decoded?