As augmented reality headsets, autonomous vehicles, and drones become more common, they will need to operate safely alongside people, and therefore have to understand the physical world with more detail than ever before.
To create these kinds of capabilities, London-based Scape Technologies has been working to create a new AR infrastructure — a map designed from the ground up for machines that allows devices to precisely determine where they are and what’s around them using only a camera.
Founded in 2016, the startup has just emerged from stealth, and has announced that it raised €7 million in seed funding last year from LocalGlobe, Mosaic Ventures, Fly Ventures and Entrepreneur First, which it has used to grow its London team to 30 people.
Along with the financing round, Scape is announcing its product, the ‘Vision Engine’ —which makes its image-based location recognition possible.
The Vision Engine is a world-scale mapping pipeline that processes images and videos, resulting in a machine-readable HD map of the environment, aligned to the physical world, in which ‘images go in, and maps come out’.
Through these HD maps, any camera can then search the Vision Engine using Scape’s ‘Visual Positioning Service’ API, to determine the device’s location without relying on expensive sensors or components.
The Visual Positioning Service is designed to be able to detect a device’s location with centimetre-level accuracy, determine a devices location within three seconds, be scalable at the city level, as well as secure and GDPR compliant.
The startup expects its technology to have applications for large-scale augmented reality, e-scooters and dockless bikes, as well as autonomous vehicles (eliminating the need for LiDAR), and plans to work with AR and mobility companies in London in 2019 to test its technologies.