Hack: This was a stupid prank where the computer by the door would detect (by machine vision) when my wife came home, let the computer in my room know, which would greet her (even though I wasn't home).
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
src
README.md
noboy0.png
noboy1.png
noboy2.png
noboy3.png

README.md

Dumb Machine Vision Prank

A few weeks ago, my wife came home from sailing:

Image of Stefanie in home

She heard me call out "Hey Stefie, how was sailing?" and she looked around the apartment for a few minutes to find me:

Image of Stefanie in home

But I wasn't home, so she eventually went back to work, and never said anything more about it. A couple weeks later, she came home from work:

Image of Stefanie in home

She, again, heard me ask from my room "Hey Stefie, how are you? How was your meeting at work," but I was nowhere to be found:

Image of Stefanie in home

After I came home, she initially didn't say anything about it, but eventually asked if I had rigged up the computer to greet her when she came home. I couldn't deny it ;)

How I did it:

april.py is the main machine vision script. It has two rectangles -- one corresponding to the door, and one corresponding to everything else. It will trigger if there is more than some amount of motion at the door, and less than some amount of motion everywhere else. The second rectangle is there so the script won't react to e.g. lighting changes. I don't know if the second rectangle is required, but I do know that I had zero false positives or negatives with this algorithm over a fairly long period of time. To use this script, there are a few things to modify:

  • rect1 and rect2 define the region of the door, and the region of not-the-door. test_run and skip_diff should be set to true to allow you to visualize where these regions are. They should both be false otherwise.
  • a and b in the main loop are measures of motion in both regions. The thresholds may need to be changed (in my case, I used 0.2 and 0.1). There was a very large range where there were no false positives or negatives.

dothat.sh runs on the machine in my living room. It runs the machine vision program, waits 2 seconds from the event (to allow my wife to come in), begins video recording, and then notifies the other machines (the IP here needs to be changed).

doit.sh runs on the machine in my bedroom. It plays the greeting, and then record video of my wife looking for me.

The gst-launch command is a beast. All it does is record video for half an hour (so I had proof). You'll need to mess with it to get the right audio/video device, and on some machines, you may need to disable audio. See the examples in the scripts. I had help with this one from the gstreamer IRC channel.

All of this runs under Ubuntu GNU/Linux, and isn't really designed to be used by anyone not computer savvy.

Copyright (c) 2011. Piotr Mitros. All Rights Reserved for photos and text. Scripts and code may be distributed/modified under the terms of the GNU General Public License version 3 or newer.