Portable 3D Scanning With eGPUs On The Go
By
Steve Johnson
We are at a crossroads of technology and communications when it comes to 3D modeling. As the tools to scan large areas get easier and smartphones and tablets become “AR ready,” it is our job as communicators, journalists, museum curators and developers to build the content that is going to help our audiences understand the world with more context.
Previously, building a photogrammetry model at the scale of a building or a city block took days of capturing images from the ground and the sky, a fiber optic internet connection to send out thousands of high resolution photographs for processing and then a pretty powerful laptop to be able to finish and design your 3D models.
All of this was expensive and time consuming – two things we do not have an infinite amount of.
Now, with the advent of smartphone cameras and foldable drones we can capture places faster than ever – creating more relevant models and experiences that can build immersive stories.
But what changed the game for us at SeeBoundless was the introduction of portable, rugged and powerful eGPU systems like the Blackmagic eGPU Pro. Having a desktop class graphics processor paired with a MacBook Pro allows us to build low resolution 3D models in near real time, in the field, as we are capturing our 3D assets. This means we can adjust flight plans, photographs we take for our data sets and change what and how we scan all in a single day of production – saving time and money for our clients.
Once we complete a scan in the field, we then in the same day start to build a high resolution textured 3D model on location.
Using desktop class graphics processors and photogrammetry software, we can have a large-scale 3D model ready for finishing and publication within 24 hours. For smaller models, say a piece or art or archaeological artifact, we can have these 3D models ready within hours.
Photogrammetry is an exact science, it’s the process in which you “scan” or photograph a place/object, how you process it and under which parameters and then using 3D modeling tools to crop, correct and finish your object. What is not an exact science is the applications in which you use 3D models to communicate, educate and build immersive stories that can help people understand new things with context. How we use this technology to highlight illegal deforestation in the Amazon, show how cities are solving problems to build a better future or bringing some of the world’s most important cultural artifacts to your living room is what makes this field incredibly exciting and rewarding.
We have the tools to efficiently scan and build 3D models for augmented reality applications and the distribution models through smartphones, tablets, 3D viewers and physical spaces like interactive exhibits to help bring the world a little closer to our audiences.
3D modeling and immersive content will soon be a cornerstone of journalism, education and museum curation. Using these models, with added context of animation, audio guides and text to build a story in which a user can move through will create a learning experience that is both impactful but also relatable.
Think about the ability to bring a life-size redwood tree into a classroom where students can see just how big its diameter is – measured not just in feet, but in relative space to their classmates, desk and classroom.
It’s the context of 3D modeling and augmented experiences that excites me the most. The context of scale to a relatable place for our users. Something we could only guess in the past, building stories with phrases like “equivalent to 1,000 Olympic size swimming pools,” can now be built with to-scale models showing just how large or small a place or object is.
We have a lot of work to do to continue to lower the barriers to this technology, but the work on the backend of processing and scanning has already made this an incredibly accessible tool to work with and I’m excited for a future when this commissioned work can be easily integrated into more work that makes a positive impact on how we learn.
Steve Johnson is Founder of SeeBoundless, a photojournalist, technologist, professor and entrepreneur leading the charge in immersive storytelling across news organizations, nonprofits and universities. Working with fellow journalists to write the manual on immersive storytelling, he has developed standards and practices, ethics guidelines and tools that are used among news organizations and universities nation-wide. Before creating SeeBoundless in 2015, Steve taught journalism at The University of Florida and was a freelance photojournalist for The New York Times, The Miami Herald, Reuters, The Orlando Sentinel and ESPN.
