[Webinar highlights] - Map Smarter, with AI-Powered MapScale®!

Last month, we hosted a webinar introducing MapScale®, a world-first and radically new way of doing indoor mapping using machine learning and smart algorithms. MapScale® enables smart buildings and location services by taking raw floorplans and instantly transforming them into beautiful, interactive 3D maps that can be edited and managed on the go in the Pointr Cloud via the Pointr Dashboard.

We’ve handpicked some of our standout webinar highlights to help you understand how you can make your buildings smart with MapScale®.

Table of Contents



Smart buildings today

Eva Cheng, Pointr’s Product Marketing Manager, discussed how the pandemic had accelerated the shift toward smart buildings, particularly over the last 2 years. Smart buildings underpinned this digital shift across verticals including smart workplaces, large healthcare organizations, the retail space, airports, and multiple buildings within campuses.

Back to top

How do you make a building smart?

A digital map was the foundation of a smart building. Digital mapping per se was not that difficult, but mapping at scale was hugely challenging. Digital indoor mapping remained a manual process, while satellites could be used to map millions of square feet outdoors fairly quickly.

Back to top

Mapping at scale required a digital solution

Manual mapping was not practical for huge companies that needed to map hundreds or thousands of buildings and offices quickly to deliver good location services and great user experiences. It became necessary to automate the process with a technology solution such as MapScale®.

Back to top

MapScale® - a world-first AI mapping technology

Pointr’s MapScale® automated the mapping production process and could quickly digitize buildings at scale, hundreds of times faster than competing vendors, and in just 2 steps.

Back to top

End-to-end solutions

Pointr offered complete solutions covering all aspects of mapping including map production, CMS, add-ons, blue dot, and more…

Back to top

The challenges of the map digitization process

Can Tunca, Senior R&D Manager at Pointr, introduced different types of floorplans, usually produced without standardization and by individuals in AutoCAD, which made it difficult to automate the process. Maps drawn in this way were not designed to be consumer-friendly and contained a host of information that wasn’t relevant to the digital mapping process.

Back to top

Simple, visually pleasing, easy-to-view consumer maps

It was important to clean up irrelevant information you didn’t want to show to the end-user for an optimal mapping experience. Certain levels of detail were also unnecessary, such as the huge variety of furniture drawn in different styles, shapes, and forms.

Back to top

Even walls were not so simple

Optimally representing walls in the final map was extremely important, and again the level of detail in drawings was often unnecessary and not required to be represented in the final digital map. Curved walls also represented a challenge as not all walls were simple geometric shapes.

Back to top

MapScale® is inspired by the human visual system

Humans could perceive stylistic differences in drawings and decide which elements were relevant and should be included in maps. MapScale® attempted to mimic the human visual system to make similar decisions.

Back to top

MapScale® in action

Melih Peker, Machine Learning Engineer at Pointr, showed a digital map (generated live from a CAD drawing) where MapScale® not only rendered the digital map but also identified furniture, rooms, elevators, stairs, and other points of interest represented in 3D. This illustrated the power of MapScale® in being able to mimic a human interpretation of the output.

Back to top

MapScale® - self-improving mechanism by design

Instead of building a huge library of rules for MapScale® (that would be inefficient and not work for every venue), we created state-of-the-art machine learning algorithms to mimic the human visual system and interpret venues as humans do. These models evolved and improved over time; we called this adaptive recognition.

Back to top

Interactive 3D maps - the power of post-processing

Once MapScale® completed the identification process, identifying items such as furniture, and classifying objects and rooms, including exact coordinates, post-processing took place. Post-processing developed the exact shape of objects such as desks, walls, rooms, etc.

Back to top

Well-rendered, integration-ready digital maps

After post-processing was complete, MapScale® then generated the finished 3D map in GeoJson format ready to be imported to the Pointr Cloud for easy management. Generated maps were also ready to integrate with major map providers such as Google Maps, Bing Maps, and Microsoft Azure maps.

Back to top

MapScale® - just the first link in a complete end-to-end product

Can Tunca illustrated that Pointr’s digital mapping didn’t end with the MapScale® generated output. Maps exported to Pointr Cloud could be edited in the Pointr Dashboard to reflect changes in buildings, the movement of furniture, room number changes, and even changes in floors.

Back to top

Metadata supported by Pointr Cloud

Metadata such as descriptions, search keywords, and your own external identifiers (your own IDs) were all supported by Pointr Cloud. Advanced configuration meant you could add extra data in a format of your choice.

Back to top

Pointr Maps on mobile

Pointr Maps could be used on desktop and mobile. Here Can showed how our maps looked on mobile - the same 3D map format as in the Pointr Dashboard. He illustrated the mobile map was updated, in real-time, with an edit he just made in the Pointr Dashboard being reflected immediately. The mobile app was fully interactive,

Back to top

Maps were highly customizable to match your brand

You could change everything relative to a map including colors, styles of objects, colors of walls, heights of walls, and much more. Present your maps exactly as you’d like them to look, even convert from 3D to 2D if you’d like.

Pointr’s 3D vector maps performed smoothly across multiple platforms and, being vector-based, flowed smoothly without glitches.

Back to top

A re-cap of Pointr’s end-to-end mapping platform

Can summed up the complete Pointr offering from MapScale’s® ultra-fast map production to creating and managing map content in Pointr Cloud and the final interactive maps supporting iOS, Android, kiosks, and web applications.

Back to top

Case study - MapScale® used in retail

Eva presented a real-life case study of how MapScale® helped one of the world’s largest retailers digitize its 2,000+ stores to help shoppers find items in-store. Retail stores changed all the time, and MapScale® was the ideal solution to keep up with this by supplying up-to-date maps to guide shoppers to multiple items around the store as efficiently as possible.

Back to top

Starting your journey with digital maps and MapScale® 

Eva talked you through what you need to get started on your journey including stakeholder buy-in, building your case, choosing your approach, identifying your requirements, and starting a pilot program. Let’s start a conversation!

Back to top

Want to learn more?
Watch the full MapScale® webinar to explore the revolutionary mapping platform that enables 100,000x faster map production.

Watch MapScale® webinar on-demand

You might be also interested in:

Contributor: Eva Cheng

Recent Posts About Mapping