From Pokémon GO to virtual AI: Niantic Spatial unveils its global 3D mapping platform

Niantic Spatial, the spin-off of Pokémon GO maker Niantic, is launching an upgraded version of its Scaniverse platform for businesses and a new virtual reality server (VPS), in its biggest release since becoming an independent company last year.
The San Francisco-based company, which maintains an engineering team of about 30 people in Bellevue, Wash., developed key components of the release in a local Seattle office, including the latest version of its developer toolkit. Niantic previously operated a key Bellevue engineering office focused on Pokémon GO and its core AR platform.
Scaniverse allows users to capture virtual spaces with a smartphone or 360-degree camera and generate detailed 3D models. The VPS, trained in part on millions of images streamed from Pokémon GO players, pinpoints the device’s location and orientation using visual cues rather than GPS.
Applications include robotics, augmented reality, manufacturing and industrial testing.
“When we came out, there was still a lot of purpose built for augmented reality and augmented reality games, even though there were a lot of requests,” said Tory Smith, director of product management at Niantic Spatial. Physical AI creates a “new head of the ocean.”
Niantic sold its games business, including Pokémon GO, to Scopely for $3.5 billion last year and spun off its Geospatial AI division as Niantic Spatial, led by CEO John Hanke, and funded with $250 million.
History and privacy: The location technology at the core of the advanced platform tracks back to data collected from Pokémon GO and Ingress players, who opted in to submit scans of real-world locations for in-game rewards.
An MIT Technology Review report last month detailed how Niantic Spatial trained its virtual positioning system using 30 billion photos taken from Pokémon GO players, raising questions about whether gamers understand how their data will ultimately be used.
Smith said data collection has always been opt-in, with no background collection, and that all information has been anonymized with GDPR-level protections in place around the world.
“I would like to put to rest any rumors that there was any secret data collection or anything like that on any of our products,” he said in an interview with GeekWire this week.
How does this work: The company’s visual positioning system determines the device’s exact location and position using what the camera sees, rather than relying on GPS. For mapped areas, it delivers centimeter-level accuracy. The new version, VPS 2.0, extends coverage to the entire world, using additional data sources to correct GPS errors even in unscanned areas.
Google provides similar virtual positioning through the ARCore Geospatial API, built into its main Street View database. Smith said the main difference is that Niantic Spatial allows customers to bring their data into space, enabling higher fidelity in indoor and private areas where Google’s public Street View can’t reach.
Customers can also keep their maps updated over time, which Google does not support.
Availability: Scaniverse will be offered with a free tier for basic scanning and positioning, with paid subscription plans that add premium features such as 360-degree camera support.
The company said it plans to add semantic understanding capabilities later this year, enabling AI to think about things and places it encounters, not just walk through them.

