Google has updated the design of its Street View cameras for the first time in eight years.
The new system will not only provide higher quality imagery for users to browse through, but will also enable image recognition algorithms to identify and collect more information to add to Google’s databases.
While over 80 billion photos have been taken in 85 countries since Street View was launched in 2007, Google’s vice president of engineering and product management Jen Fitzpatrick told Wired that user interactions with Street View have become more complicated.
“People are coming to us every day with harder and deeper questions,” Fitzpatrick said.
Rather than simply searching for a images of familiar street, Fitzpatrick said that users are now requesting more complex information that might involve finding a local restaurant by its opening hours or a business by the colour of the building it is in.
“These are questions we can only answer if we have richer and deeper information,” Fitzpatrick said.
Machine learning and artificial intelligence are used to gather this information, which will become more reliable thanks to the higher resolution images of the new Street View system.
Two cameras mounted on either side are able to capture high definition photos of buildings and street signs, and can potentially identify things like business names and opening hours that can be used across Google’s products.
Street View can also add new addresses to its maps by identifying street names and numbers.
Google said that over 10 million miles, or 16 million kilometres, of panoramic imagery has been captured since 2007.
