When a person takes a shower, it is often the cast that some fraction of the water leaving the shower head lands directly on the shower floor without hitting the shower-taker (Figure 1).
This “off-target” water is totally wasted.
Thanks to advances in computer vision and inexpensive electronics, it is now possible to figure out, in real time, where the water that leaves each shower nozzle will actually end up. Specifically, we need to determine if there is a human in the path of the water.
In order to accomplish this, a valve is added to each individual shower nozzle (allowing it to be independently opened and closed) and a camera is mounted on the shower head to allow it to track the shower-taker (Figure 2).
The camera feed from the shower is sent to a secure cloud facility, where an advanced machine learning algorithm analyzes it to determine which shower nozzles should be opened or closed. This system should be operable with latency of less than one-tenth of a second, which should be more than sufficient..
This system sells itself: it saves water without negatively impacting the shower-taking experience.
PROS: Saves water and improves shower water pressure.
CONS: Some people might object to having a video feed of themselves in the shower being constantly streamed over the Internet, but it is very unlikely that an Internet company would have a security breach.