>What did you mean by "WebHook was updated to indicate if a person caused the motion"?
I thought I read somewhere that motion based webhooks had been updated to include info about weather the motion includes a person or was triggered by a person. However I can't find that info now, so that now that looks doubtful.
I'm worried about the scalability of the webhook system.
I'm currently doing a trial with a retail store with around 20 cameras. I was hoping to use MQTT with the snapshot API, but I wont be able to get it to scale. I already exceed the 5 API calls per second and once the queue starts growing it looses all resemblence of real time.
I'm worried that if the trial is a success and it gets rolled out to the other stores I could be needing thousands of webhook motion detections being required per second.
The Meraki MV Sense system (with regard to MQTT and real time imaging processing) seems good if you just have one camera, or are just playing with it in your office, but the architecture just doesn't allowed it to be scaled up into larger commercial scenarios.