Hi everyone,
I was wondering if anyone has tried or might have. some advice or ideas on how to implement this model on the OAK-D:
'person-detection-action-recognition-0006.blob'

I found it on the blob converter website http://69.164.214.171:8083/
It is all still a little complicated for me, so I hoped someone could steer me in the right direction on how this model could be used on the device.

ps: Thank you all for this community, I am quite new to these technologies so its amazing to see how everyone is really willing to help. Makes it so much easier to get started!

5 days later

Hello @CorLangerak ,
so first you would have to download the blob. I used http://69.164.214.171:8084/ to download the
person-detection-action-recognition-0006.blob. After that, I used this simple script:

Note that decoding is not yet finished. From the openvino docs you can see that the output is a bit complicated to decode. If you get it to work feel free to share it with us🙂
Thanks, Erik

    2 years later

    erik Hi, is there any work done on this model output decoding? Finished or Not!?

    • erik replied to this.
      4 days later

      Hi SamiUddin , Do you mean on-device model decoding? I don't think so, but you could utilize Script node for that.