Latest Invention: Program that Identifies and Erases People from Google Street View

Although Google Street View is useful and entertaining tool, some people are skeptic about it mainly because they don't want to appear on the images taken by the Street View team, especially when there is a risk that some photos may end up on the Internet.

In order to get rid of the problem, a graduate student from the University of California, San Diego, decided to invent software that can digitally remove people spotted on the street by Google Street View team.

Currently the program does not have a name, but Arturo Flores developed it as a proof-of-concept for CSE 190A. Despite the fact that Google Street View blurs the faces of the pedestrians, some consider that it is not enough since by analyzing the clothes, body shape and approximate location it is still possible to identify the individual on the picture.

The latest invention from Mr. Flores is able to identify human forms in each picture, erase these forms and fill the blank space with background pictures that were taken before or after.

However, there are several drawbacks of the system. For instance, there can be problems with erasing a human form when a person walked the same way as the Street View camera car at such a speed that the individual simply blocked the same part of background for a number of shots that were taken in a row, reports Gizmag.

Beside, Mr. Flores' invention only functions in urban settings, because the backgrounds there are usually flat.

When the graduate student started working on the system, he used the Street View pedestrian locator that was developed by computer science professor Bastian Leibe from Aachen University. Currently Mr. Flores is working on improving the software so it could identify and remove groups of people.