01·3D & Spatial
3DGS Research Pipeline
26 Gaussian-splat engines integrated into one benchmark harness, evaluation in progress. Which one actually survives a real luxury-interior scan, not just the leaderboard.
ResultA reproducible benchmark harness: 26 engines wired, a scan-difficulty taxonomy, and per-engine failure modes, feeding the academic papers now in progress.
[ Ask about this build ]01Discover
Why it exists.
The brief said nothing about Gaussian splatting. It said find a better way to show property than Matterport, which every brokerage already had, and which made every brokerage's tours look like every other brokerage's tours. The answer was not another capture vendor, it was that the underlying method had changed: 3DGS.
3D Gaussian Splatting moved fast between 2023 and 2026, with new engines shipping every quarter, each claiming higher PSNR on its own benchmark scenes.
This pipeline is built to answer it: 26 Gaussian-splat engines wired into one harness, evaluation in progress across scans from a clean studio booth through a mirror-walled marble lobby to a city-scale neighbourhood capture.
The pipeline also covers the unbuilt case: novel-view synthesis via video diffusion, monocular depth lift, and the hybrid stack that bolts those onto a real scan when only half a building has been constructed.
02Define
The brief.
Answer which Gaussian-splat engine actually survives a luxury real-estate brief on a real scanner, not which one wins average PSNR on clean studio scenes.
03Develop
What it took
Skills behind it.
Primary discipline plus the support stack.
Skills demonstrated
- ResearchPrimary
One harness spanning studio scans to city-scale capture; PSNR/SSIM/LPIPS evaluation and the full matrix are still ahead.
- 3D & SpatialSecondary
Novel-view synthesis and monocular depth paths for unbuilt properties stitched into the same pipeline.
- AI & Machine LearningSecondary
Video-diffusion novel-view models evaluated against COLMAP-grounded baselines.
- Data PipelinesSupporting
The eval harness itself: per-engine launch scripts, logged failure modes, reproducible runs.
Develop · Notes
The build, in full.
How it came together and the decisions on the record - kept off this page so the case study stays a read, not a scroll.
Read the full build notes04Deliver
What shipped.
A reproducible benchmark harness: 26 engines wired, a scan-difficulty taxonomy, and per-engine failure modes, feeding the academic papers now in progress. The full engines-by-scans matrix is still ahead. A relighting track runs alongside it: an interactive viewer with movable light fixtures and real soft cast shadows, plus delighting and relightable retraining still in progress.
The write-up is finished, four pages and anonymised for double-blind review, though not yet submitted. The eval framework keeps running against new engines as they ship, and adding one is roughly a day's work now. Early results already shape which engine gets tried behind Barnes Vantage, ahead of the full matrix.
By the numbers
Engines integrated
Evaluation matrix
PSNR/SSIM/LPIPS
Research origin
The paper-to-product thread this project sits inside. The full chain, paper through benchmark to shipped, is on the thread.
- 3D Gaussian SplattingSee thread
Turning ordinary photographs of a real place into a 3D scene you can walk through in a browser.
- Mirrors and glass captureSee thread
The open problem Gaussian Splatting still hasn't solved for high-end interiors.
See also
Research
Academic Papers
Two companion academic papers on applied Gaussian splatting for luxury interiors, arguing robustness matters more than leaderboard PSNR.
3D & Spatial
Barnes Vantage
An international portal for viewing real Dubai residences in 3D: walk a real scan yourself, or have a broker drive the same walkthrough live.
3D & Spatial
Scan to Floor Plan
Turning a 3D scan into a drawn-to-scale floor plan, and the measured finding that the machine-learning version lost to a far simpler method.