Automating your home with computer vision

[miguelaeh (Miguel Ángel Cabrera Miñagorri)]

Hi!

Tutorial/demo: https://youtu.be/4xVDB5DIC7k?feature=shared

I recently created an integration/custom component for Pipeless Agents, which allows you to automate your devices using your cameras and computer vision.
It does not require knowledge of computer vision since you can create custom filters just with natural language. You can also provide a Python script that receives structured data from the vision filter, so you can easily create your own complex workflows.

I created a video tutorial on how to set up a small project to turn off the TV when the person watching it leaves I think the possibilities are huge since we go beyond just detecting motion to actually detect anything, even with the context of what happened before.

[kirkfrey (kirk)]

just ran across this. so I have an idea but no clue how to execute it. As I was doing the “dog doo search and recover” while they were playing I realized - I have a camera out here and wouldn’t it be great if the camera could record the general position of the dog when he/she “goes” and then mark that on a picture/map? This would keep me from having to execute the grid search pattern of the entire backyard. I’ve no clue how to even start something like this.

thoughts?