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 165 forks source link

Segmentation fault when running object_detection_demo_yolov3_async with custom model #55

Open srinikethms opened 5 years ago

srinikethms commented 5 years ago

[Required] Your device (RaspberryPi3, LaptopPC, or other device name):
RaspberryPi3 [Required] Your device's CPU architecture (armv7l, x86_64, or other architecture name):
armv71 [Required] Your OS (Raspbian, Ubuntu1604, or other os name):
Raspbian [Required] Details of the work you did before the problem occurred:
Compiled the c++ program as instructed and tried to run a custom tiny-yolov3 model. [Required] Error message:
segmentation_fault

[Required] Overview of problems and questions:

How to fix the segmentation fault ? Is it due to the custom model ?

PINTO0309 commented 5 years ago

I don't know the structure of the model you created, so it is difficult to answer.

srinikethms commented 4 years ago

I don't know the structure of the model you created, so it is difficult to answer.

Thank you for your prompt reply.

I have actually retrained the tiny yolov3 model with custom data and to have 2 classes instead of 80 classes.

Also, I was able to successfully compile and run your asynchronous code with my custom model in Ubuntu desktop, but I am still unable to run the compiled code in Rpi3. And, I am not able to run even the default models (tiny-yolov3 and yolov3) with the compiled code in Rpi3.

Can you please advise me, if I need to configure anything in RPi in specific to run the compiled code with the yolov3 models ?