TIER 1 — READY NOW5 MODULES5 INTEGRATED MODELS~25 FPS REAL-TIME

PHANTOMAGIA

The Magic 3D Room Scanner

Point a phone at a room. Walk through it for 60 seconds. Get a complete 3D digital twin where every object is labeled, measured, and individually extractable.

TOTAL ADDRESSABLE MARKET

$50B

TIER 1 // 01

RGB Video → CHRONOS → PROTEUS + LOGOS → PLEROMA → NEXUS

01THE PROBLEM

3D SCANNING IS BROKEN

Matterport charges $3K+ per scan with $50K+ hardware. Apple's RoomPlan needs LiDAR and produces toy-quality output. Google's ARCore gives you point clouds with no semantic meaning — just dots in space.

$3K+
COST PER SCAN TODAY: $50K+
HARDWARE SETUP: 0
SEMANTIC UNDERSTANDING

02THE SOLUTION

FIVE PAPERS. ONE PIPELINE. ONE CAMERA.

PHANTOMAGIA chains five research breakthroughs into a single real-time system. RGB video in, structured 3D understanding out. No LiDAR. No photogrammetry expertise. Just a camera and a walk.

MODULE CHAIN:RGB Video → CHRONOS → PROTEUS + LOGOS → PLEROMA → NEXUS

01
CHRONOS
Temporally Consistent Depth
Smooth, stable depth fields where surfaces stay locked across hundreds of frames. No flicker. No jitter.
02
PROTEUS
Persistent Object Tracking
SAM 2.1 instance segmentation with memory. Every object gets a persistent mask that tracks through the entire video.
03
LOGOS
Language Grounding
Not "segment #47" but "red ceramic mug, 8cm tall, second shelf from top." Makes output actionable for robots and humans.
04
PLEROMA
Per-Object 3D Reconstruction
Each tracked segment becomes a textured, extractable 3D model via Gaussian splatting. The room is a scene graph of objects.
05
NEXUS
Semantic Exploration
Identifies unscanned regions using semantic ray frontiers. "There's more behind that bookshelf — go look."

03WHAT THIS MEANS

FOR ROBOTICS

Every robot that operates indoors needs a 3D world model. PHANTOMAGIA makes 3D world models disposable. Scan a new warehouse in 10 minutes. A shelf moved? Re-scan that aisle in 30 seconds. New facility? The robot builds its own map on arrival.

01Collapses deployment cost from weeks of setup to hours
02Makes home robots viable — no LiDAR technician needed
03Every robot becomes spatial-first when 3D understanding is cheap and instant

04NUMBERS

KEY NUMBERS

25FPS
END-TO-END REAL-TIME: 5MODELS
INTEGRATED CHAIN: 1CAMERA
SINGLE RGB INPUT: $50B
ADDRESSABLE MARKET

05MARKET

WHERE THE VALUE IS

WHERE THE VALUE IS
E-COMMERCE FULFILLMENT	$25B
REAL ESTATE VIRTUAL TOURS	$12B
INSURANCE DAMAGE ASSESSMENT	$8B
INTERIOR DESIGN	$5B

THE MOAT

Five models chained into one pipeline. The industry has depth estimators. The industry has segmenters. The industry has 3D reconstruction. Nobody connects temporal depth to semantic tracking to per-object mesh to exploration planning in a single real-time system.

The chain is the moat.