Visualizing Locations for a Search Engine for the Physical World.

Today, search engines on the Internet allow us to instantly find almost any kind of information or media. Unfortunately, the same capabilities are not available in the physical world, except for in industrial settings such as automated warehouses. Therefore, we implemented a stationary prototype that supports finding objects in everyday environments. The location is represented using a web-based search interface. In a user study, we compared four different representations for our stationary object finding system. Results favor a last seen image representation and a 2d map and show that more complex representations are not suitable for representing results of a search engine for the physical world. The participants were throughout positive about using a system like this in their daily lives, although there were concerns for privacy. Based on these results, we argue that scalable physical search engines can be built for everyday environments; that searching for physical objects can be made understandable to users; and that there are many practical and commercial cases where such systems would be beneficial.