Mapping Startup Mapillary Hires Computer Vision Expert From Apple

author: Eric Walz   

Sweden-based mapping startup Mapillary has poached a high-level computer vision expert from Apple, the company said on Wednesday.

Till Quack, an engineering manager who oversaw augmented reality (AR) and self-driving efforts for Apple, including the company's secretive autonomous driving program, will become Mapillary's vice president of product, the company told Reuters.  

Prior to joining Apple, Quack worked for chipmaker Qualcomm. Quack will be responsible for the important task of keeping Mapillary's maps updated.

Mapillary was founded in 2013, after CEO Jan Erik Solem, also a former Apple employee, sold his facial recognition startup to Apple. The company has a unique approach to map building.

Mapillary uses crowd-sourced data from a global network of contributors to collect street-level images. These images are collected from volunteers and organizations using smartphones or in-vehicle dash-mounted cameras.

Then, using computer vision techniques, Mapillary stitches these images together to create immersive, high-definition maps that include street-level views similar to Google Maps Street View. The technology can identify road signs, buildings or other infrastructure, including trees and traffic signals. These high-level details are added to the maps.

"The main product that we build is the platform to convert imagery from anyone into map data we can publish for everyone," Solem said to Reuters.

Keeping maps updated is a huge challenge, as road frequently change due to construction zones, lane realignments, as well as new roads being built. For drivers or autonomous driving vehicles, having access to current maps is vital for precise navigation.

Tech giants Google and Apple solve this problem by sending out fleets of their own vehicles outfitted with cameras and other sensors to gather images of the roads, so their maps stay updated.

For example, if a new lane was added to a section of road, Google's own mapping vehicles would need to traverse this section of road to gather the necessary data Google requires to update its maps. Mapillary can do this task much quicker using crowdsourced data.

Mapillary's software can take pictures from nearly any kind of camera to build robust, three-dimensional maps.

The company provides this data to mapping firms, such as HERE Technologies, the mapping firm owned by a consortium of German automakers including Daimler, BMW and Audi. These maps will eventually be used for autonomous driving.

Mapillary is already experimenting with using the low-resolution cameras that are already built into many vehicles for features such as lane keeping assist, to help update its high-resolution maps.

Other startups including Mapbox, Civil Maps, DeepMap are all working on mapping technology and together raised more than $270 million in venture capital.

Mapillary has raised $24.5 million from investors such as Sequoia Capital and BMW Group's iVentures.

Eric Walz

Originally hailing from New Jersey, Eric is a automotive & technology reporter covering the high-tech industry here in Silicon Valley. He has over 15 years of automotive experience and a bachelors degree in computer science. These skills, combined with technical writing and news reporting, allows him to fully understand and identify new and innovative technologies in the auto industry and beyond. He has worked at Uber on self-driving cars and as a technical writer, helping people to understand and work with technology.