While I run "python3 ./stream_infer.py " on the raspberrypi, I got this wrong. Ask for help!!!

[idata]

pi@raspberrypi:~/ncapi/py_examples/stream_infer $ python3 ./stream_infer.py

** (stream_infer.py:2841): WARNING **: Error retrieving accessibility bus address: org.freedesktop.DBus.Error.ServiceUnknown: The name org.a11y.Bus was not provided by any .service files

Number of categories: 1000Device 0 Address: 1.2 - VID/PID 03e7:2150Starting wait for connect with 2000ms timeoutFound Address: 1.2 - VID/PID 03e7:2150Found EP 0x81 : max packet size is 512 bytesFound EP 0x01 : max packet size is 512 bytesFound and opened devicePerforming bulk write of 825136 bytes…Successfully sent 825136 bytes of data in 46.907734 ms (16.775720 MB/s)Boot successful, device address 1.2Found Address: 1.2 - VID/PID 040e:f63bdoneBooted 1.2 -> VSCSegmentation fault

[idata]

when I do the comment "sudo apt-get install at-spi2-core" .

I can Solve the problem ----------------- ** (stream_infer.py:2841): WARNING **: Error retrieving accessibility bus address: org.freedesktop.DBus.Error.ServiceUnknown: The name org.a11y.Bus was not provided by any .service files

but there are also have problem , because i can not get the Camera stream. and get this result

pi@raspberrypi:~/ncapi/py_examples/stream_infer $ python3 ./stream_infer.py

Number of categories: 1000Device 0 Address: 1.2 - VID/PID 03e7:2150Starting wait for connect with 2000ms timeoutFound Address: 1.2 - VID/PID 03e7:2150Found EP 0x81 : max packet size is 512 bytesFound EP 0x01 : max packet size is 512 bytesFound and opened devicePerforming bulk write of 825136 bytes…Successfully sent 825136 bytes of data in 46.936681 ms (16.765374 MB/s)Boot successful, device address 1.2Found Address: 1.2 - VID/PID 040e:f63bdoneBooted 1.2 -> VSCSegmentation faultpi@raspberrypi:~/ncapi/py_examples/stream_infer $

[idata]

export NO_AT_BRIDGE=1

[idata]

I have try run "export NO_AT_BRIDGE=1"comment , but it does not work.

[idata]

@tiandixing Are you using a Raspberry Pi Camera or a USB camera? This particular example (stream_infer) uses a USB web camera.

[idata]

@tiandixing, if you are trying to perform image classification on live camera feed using RPi camera module, you can try one of the following:

1. Modify stream_infer to use the PiCamera API
2. Enable bcm2835-v4l2 driver so that RPi camera shows up as a V4L2 device.

Here's a blog post that uses live-image-classifier to run image classification on RPi Cam's live feed - https://movidius.github.io/blog/battery-powered-dl-engine/

[idata]

@Tome_at_Intel I used a USB camera .

while I use sudo python3 ./stream_infer.py , "Segmentation fault"had been Solved, but I can not open the GUI .

pi@raspberrypi:~/ncapi/py_examples/stream_infer $ python3 ./stream_infer.py

Number of categories: 1000Device 0 Address: 1.2 - VID/PID 03e7:2150Starting wait for connect with 2000ms timeoutFound Address: 1.2 - VID/PID 03e7:2150Found EP 0x81 : max packet size is 512 bytesFound EP 0x01 : max packet size is 512 bytesFound and opened devicePerforming bulk write of 825136 bytes…Successfully sent 825136 bytes of data in 46.936681 ms (16.765374 MB/s)Boot successful, device address 1.2Found Address: 1.2 - VID/PID 040e:f63bdoneBooted 1.2 -> VSC

I want try the example like this :smile:

https://ncsforum.movidius.com/discussion/118/movidius-nc-sdk-1-07-07-with-raspberry-pi-support

ASK FOR HELP !!! @Tome_at_Intel

[idata]

the GUI is opened,and quickly shut down automaticly. can not get the Video stream。@Tome_at_Intel

[idata]

@tiandixing Make sure you have installed the gstreamer 1.0 python dependencies by using the following command sudo apt-get install python-gst-1.0 gstreamer1.0-plugins-bad gstreamer1.0-libav. If you are getting a Myriad Error, please double check your graph files to ensure that the NCSDK version on your host machine and on your RPi are the same and recompile the graph file with mvNCCompile.

[idata]

@AshwinVijayakumar We used your method, while we run "image-classIfie.py" ,we got error

usr/local/lib/python3.4/dist-packages/mvnc/mvncapi.py line264,inGetResult raise Exception(Status(status)) Exception:mvncStatus.MYRIAD_ERROR

mvnccompile command not found

[idata]

@AshwinVijayakumar could you tell us how to install opencv3 in raspberrypi 3 B, when we run "import cv2" under python 3 , It tells wrong. Thank you very much!

[idata]

@tiandixing You can check out this site on instructions for how to install OpenCV on the Raspberry Pi.

[idata]

could you tell us how to install opencv3 in raspberrypi 3 B, when we run "import cv2" under python 3 , It tells wrong

OpenCV should have been installed as part of your NCSDK installtion (unless you chose not to, or you installed API only). However, you can install OpenCV by running https://github.com/movidius/ncsdk/blob/master/install-opencv.sh on your RPi. Note that there is no pip3 wheel for RPi, so the script will build & install from source - it takes quite a while to do this.

usr/local/lib/python3.4/dist-packages/mvnc/mvncapi.py line264,inGetResult raise Exception(Status(status)) Exception:mvncStatus.MYRIAD_ERROR mvnccompile command not found

The Exception:mvncStatus.MYRIAD_ERROR seems to be an issue with the API unable to access the NCS hardware; try unplugging the device, re-plug and re-run the program. I don't understand the relation of mvnccompile command not found with the first error. Can you please paste the full console output?

[idata]

SINK_NAME="ximagesink"

Try modifying pipeline by adding videoconvert element as below in stream_infer.py

view_launch_str = "\

queue max-size-buffers=2 leaky=downstream ! videoconvert !\ %s name=%s" % (SINK_NAME, GST_VIEW_NAME)