A startup out of Sweden called Mapillary is using crowdsourced photos to create an open and more intelligent version of Google’s Street View, and today it announced a Series A round of funding from an impressive list of backers to help make that ambition a reality.
The $8 million from Atomico, Sequoia, LDV Capital, and PlayFair will be used to expand operations outside of its home base of Malmo, including a new office in San Francisco, and also to hire more staff, specifically computer vision engineers to supercharge the kind of data that can been gleaned from Mapillary’s image library.
“The vision of Mapillary is to map the world with photos,” founder Jan Erik Solem told me in an interview earlier, “but behind that, the ambition is to interpret as much as we can from these photos so that you can figure out what’s happening in places, so that you can make intelligent decisions about places. We still have a lot of work to do.”
But if there is a founder fit to take on Google in mapping and image databases, Solem is a very strong pick. His last company, the facial recognition company Polar Rose, was acquired by Apple, where he then stayed for several years doing research into computer vision and other areas.
The founder pedigree was also one of the big reasons behind investor interest in Mapillary, too.
“What we are very excited about there is the team. Computer vision is something they know very well,” said Niklas Zenntrom, co-founder of Atomico. “Also what we see here is that the world is becoming more dependent on this type of imagery, there are so many different services that need street-level imagery that is up to date, and we believe that the community-based, crowdsourced model could have a much higher frequencies of uptake compared to Google. We think that what they are building up is an asset that becomes more valuable as the data becomes richer.”
Mapillary announced Seed funding about a year ago (Sequoia led that round), and in the last year it has been building up a community of users globally, some 18,000 photo enthusiasts who have uploaded photos to contribute to a database that now covers over 1 million miles and more than 50 million images.
These are matched up with maps from Mapbox, and users can search images and get exact map locations, or reverse search map points to find photos. These can be viewed separately, or they are also collated together in clever, “3D” video-like passages (you can see examples on Mapillary’s homepage).
But as you can probably guess, this is just the tip of the iceberg for what is out there and what can be covered… and what Google has tracked through its own Street View investments and those images that Google itself crowdsources.
Solem tells me that while Mapillary is intent on building up that user community and the database that it produces, this is not the startup’s primary focus for making money. He says there are no plans to introduce any kind of advertising or other forms of monetization around the consumer-facing service.
Instead, the company is focused on B2B, specifically providing a platform for governments, other organizations and businesses to use and upload images through a variety of formats to use them for their own purposes. These pictures will become part of Mapillary’s database, but they will also be able to be “read” for the purposes of the organization or project in question.
One example Solem provided for where this could be useful is in public works.
“Everything you see in the real world, organizations are responsible for maintaining it, be it roads or signage or making sure street lights work or trees are cut,” Solem says. But many of the options available to organizations today for visually recording locations to track these kinds of things tend to be very expensive, in the order of hundreds of thousands of dollars when you hire specific companies or buy special equipment to do it yourself, he says.
Mapillary, in contrast, has built out an SaaS model priced at $200/month or $1,000/month depending on the level of data usage. Solem says there will likely be further tiers introduced that will include the usage of Mapillary’s own rigging equipment to record pictures.
Organizations that have already signed up to use the startup’s services include towns in its home market such as Helsingborg, as well as Los Angeles County, and the World Bank and the Red Cross.
Mapillary is not the first crowdsourced mapping effort to emerge. Waze, which crowdsources traffic and navigation data, is perhaps one of the biggest and most successful. What’s notable is that it is now owned by Google (who fended off acquisition interest from another tech giant, Facebook).
Despite his own history in selling to a big tech giant, and the track record of so many other mapping startups getting snapped up, Solem believes that there is not just an opportunity, but a need for standalone businesses like Mapillary to make headway in this market.
“I think it is worrisome that in mapping things are being consolidated into a few players,” he says. “Traditionally it’s been dominated by them because it’s so expensive to do anything in maps, but with the innovation in mobile there is a potential for a shift that allows smaller players to provide specific data, even if a lot of them are getting picked up, such as Waze.
“It’s bad because it means that data moves into silos and very little is shared again. When apple picks up companies and puts their data into Apple Maps, they disappear. A lot of the data that used to be provided is gone. And Apple has no interest in providing that info to anyone else. There are certain things that you should keep independent.”