Well, not so much a poor man, as one who doesn’t want a full power computer doing a task that can be made so much simpler!
I must say that, in all probability, someone has done this before, as it is elegant and simple, compared to the standard way of doing video frame comparison. However, the nearest I can find is using video tools to play with audio!
Given that most people get bored with watching a screen, waiting for something to happen on a static image, it might be easier for them to listen to the image. Demodulate the video signal, and transform it to an audio feed, put out via a speaker. For a static screen, the sound will be a constant hum, but if something moves, or the image changes, even slightly, the sound will change. Since you are only interested in the event, this would be a cue to check the screens.
Taken a step further, a relatively low-powered processing system could be \”trained\” to distinguish between things moving towards it, away from it, left-to-right, etc. and this could have applications for video watch-dogs, household robots, and fun toys for big kids!