Rethinking Cloud Strategy for Spatial Data
For decades, the web and spatial data have had an uneasy relationship. File sizes were massive, formats were inconsistent, and platforms rarely supported more than one dataset at a time. Real-time collaboration—the kind people take for granted in tools like Google Docs—was virtually impossible. As a result, professionals working with reality capture have historically relied on desktop-based software.
But desktop workflows come with their own limitations. They require high-powered machines to process and visualize dense datasets, and they inherently silo users, making cross-team collaboration difficult. Even though spatial data holds enormous potential for asset inspection, infrastructure management, and digital documentation, extracting insights has often remained the domain of highly trained specialists.
In recent years, cloud platforms have made progress in democratizing access to spatial data, but much of this innovation has focused on meshes and orthophotos. LiDAR, once too costly for widespread adoption, is now entering the mainstream. It brings new opportunities but also new hurdles:
1. Multiple formats (LAS/LAZ, E57, PLY, XYZ, and more)
2. Differences in georeferencing (drone and terrestrial versus SLAM)
3. File sizes that can range from megabytes to terabytes
For LiDAR and spatial data to truly become accessible at scale, the cloud must be reengineered to handle these challenges.
Why The Cloud Has Been Such a Pain: A USGS Case Study
Consider the U.S. Geological Survey’s 3D Elevation Program (3DEP) – one of the most ambitious public spatial data initiatives in the world. Since its launch in 2012, 3DEP has aimed to provide nationwide LiDAR coverage, contracting enterprise firms like NV5, Woolpert, and Sanborn as well as regional service providers to map hundreds of thousands of square kilometers across the country. The resulting datasets are published through the 3DEP LidarExplorer, a web portal intended to make this treasure trove of elevation data accessible to the public.
Now, suppose you’d like to access these publicly available spatial datasets from 3DEP. Here’s how you’d currently go about this.
Step One. Go to the 3DEP LidarExplorer page, draw an Area of Interest (AOI), and select your region. For this case study, let’s look at San Francisco.
After drawing your AOI, choose from available collections which, for San Francisco, include data captured in 2004, 2010, 2016, 2018, and, most recently, 2024. Let’s pull the 2024 LiDAR Point Cloud (LPC) and Source DEM datasets.
Step Two. After clicking through the data access links, you’ll be taken to an intermediary page that provides a few more links to download the LAZ and TIFF files.
Step Three. After clicking on the LAZ and TIFF links, you’ll be presented with a massive list of available datasets for download:
Step Four. There is no step four. There’s no in-browser visualization, no preview, and no immediate context. You can only download the files.
And this is where frustration sets in. Despite hundreds of millions of dollars invested and countless hours of fieldwork, most users are left staring at gigabytes of cryptically named files they can’t easily open. To find the right scene, you may have to download hundreds of tiles and check them one by one. For the average person—or even a non-specialist professional—this “public access” feels very much not accessible.
Introducing Stitch3D – A Modern Cloud for Spatial Data
So, what would an ideal cloud platform for LiDAR and spatial data actually look like? At minimum, it should:
1. Host, preview, and measure multiple point clouds as layers on a map
2. Allow real-time collaboration, as seamless as working in a Google Doc
3. Switch effortlessly between 2D and 3D views
4. Manage projects, file types, and customer interactions in one place
That’s the vision behind Stitch3D. The platform was designed to take the friction out of spatial data workflows, turning cryptic downloads into collaborative, cloud-native projects.
Let’s revisit our San Francisco case study. Instead of being stuck with a pile of LAZ and TIFF files from 3DEP, here’s how the process plays out on Stitch3D:
Step One. Create a project in Stitch3D and add your metadata and coordinate reference system (CRS) info in the “Details” tab.
Step Two – Upload up to 10 LAZ or TIFF files at a time with a 25GB limit per file. Organize them into folders for easy navigation. For this example, we brought in 50 sample LAZ and TIFF files from the 3DEP website.
Once uploaded, the files live in a unified project workspace.
Step Three – Explore your data! Click on any of the files and the Stitch3D viewer renders the point cloud and ortho/DEM files directly in the browser.
Try toggling each point cloud layer on and off to inspect one or multiple files at a time. You can also measure, annotate, and crop each LAZ file or switch over to the 2D viewer to see orthophoto and DEM files.
Within the 2D viewer, you can adjust the “opacity” slider to inspect data overlays for different file types or different capture dates/times.
And here’s the kicker: it’s multiplayer. Multiple users can inspect, annotate, and collaborate in the same 3D environment, turning what was once a solitary, desktop-bound task into a shared, cloud-based experience.
And that’s it. The cloud is now optimized for a true 3D spatial experience. Go ahead and give it a try today. Create a free account at www.stitch3d.io to upload your first project!
For any questions, please reach out to hello@stitch3d.io.
