Uploading our object detection data to website

Hello,

For my project I’m using the Jetson Nano 2GB model, we would like to upload data from our object detection to a website, we dont need to have the video feed uploaded we would just like to find a way to send data to a website to tell whether a chair is occupied or not. We currently have our model detecting empty chairs and chairs that are taken (with people in them), and we would somehow like to make a website that will tell how many chairs are available vs how many are taken. Does anyone know of a way to do this with edge or can point me in the right direction? I have not been able to find any tutorials or information as to how to do this, I would really appreciate any insight.

Our other idea, is to use Twilio like Jan did in his video “Arm your Jetson Nano with Tony Stark SuperPowers” (great video btw <3). When the Elephant and Jan’s bounding boxes intercepted each other he got a text message saying danger, or something along those lines. We were thinking maybe we could set up a Twilio service where people could put their phone numbers into our website and have the text messaging service tell them how many seats are available in the room they are going to. For example, they would get a text message saying which room number are you going to, and then Twilio would respond with, "this is how many seats are left: / This is how many seats are occupied: ".

Any help would be greatly appreciated, we’ve tried doing some Django work but not sure how that would work with edge.

Any help/guidance would be greatly appreciated!

Justin

Hi @jumill87,

You can go ahead and follow our object detection classification tutorial to build your model, then when you are done training you can follow the linux object detection Twilio tutorial to deploy the model on the device and sent alerts over Twilio.

Please let me know if you have further questions!

Hey @jenny,

Thank you for the help, hopefully we can come up with something cool. Will let you know if I have anymore Q’s, thanks again.

Justin

1 Like