Too big to open
Multi-GB point clouds take too long to download and view.
Multi-GB scans, usable on the web and in CAD
We build pipelines that stream, clean, classify and reconstruct point clouds and meshes, so scan data becomes something people can work with.
Problems it solves
Why teams bring us in
Noise and gaps
Raw scans need cleaning and repair before anyone can use them.
Points, not models
Engineers need surfaces and meshes, not just millions of points.
Hard to share
Scan data stays on the desktop of the person who captured it.
What we deliver
What you get
Delivered as a module inside your existing product, in your codebase and with your team.
Octree streaming
Level-of-detail streaming that shows detail only where it is needed.
Surface reconstruction
Watertight meshes from scan data, ready for analysis and CAD.
Semantic classification
Points labelled by type, such as ground, buildings or vegetation.
Gaussian splat rendering
Photoreal captured scenes, streamed progressively in the browser.
Case studies
Where we have delivered this
Reality capture / Survey
Streaming Large Point Clouds to the Web
An automated 3D Tiles pipeline and streaming service that renders point clouds several GB in size in the browser.
Read case study
AEC / Construction
Surface Mesh Optimization
High-quality, low-polygon surface meshes created from point cloud data for fast web rendering.
Read case study
Reality capture / Survey
3D Mesh Reconstruction with ML Segmentation and Adaptive Point Clouds
Segmented, semantically classified 3D meshes reconstructed from images, using ML segmentation and adaptive point clouds.
Read case study
Reality capture / Survey
3D Reconstruction from 360° Panoramas
An automatic tool that turns 360 color and depth panoramas into a textured, low-polygon 3D mesh.
Read case study
Technologies and formats
What we work with
Industries
Where this applies
How we deliver
From first call to long-term support
- 01
Discovery call
30 minutes with an engineer, under NDA if you need one.
- 02
Technical discovery
1 to 3 weeks: requirements workshop, code and data review, and a proof of concept for the risky parts.
- 03
Proposal
Phased plan, team, estimate and engagement model.
- 04
Delivery
Two to three week iterations, with a demo every sprint.
FAQ
Common questions
How large a dataset can you handle?
We design for multi-GB point clouds, streaming only what each view needs.
Can you produce CAD-ready output?
Yes. We reconstruct, repair and simplify meshes for downstream CAD and analysis.
Are any of these tools open source?
Yes. three-e57-loader and three-gsplat-loader are published on npm.
Need this in your product?
Tell us about the module you need. You will talk to an engineer from the first call.