TrigunAI  ·  Deep Tech

One photograph.
A rigged human
that walks.

Pratima turns a photograph into a fully-rigged 3D human, running in real time on a standalone headset — in about fifteen minutes. Not generated art. Measured geometry, checked against spec, and refused when it doesn't meet it.

13,080 PTS FACE MESH 26 JOINTS ONE SHARED RIG ~40 AUTO CHECKS ANY ONE CAN REFUSE
~15 minphoto to finished character
13,080points in the face mesh
26bones, shared by every character
~40automatic checks — any one can refuse
1 of 6pipeline stages that invent anything
§01

An algorithm,
not a model.

The natural assumption is that a neural network does the whole thing — a photo goes in, a character falls out. That's not how Pratima works.

Exactly one of six stages invents anything: a generative model that turns your photograph into a rough 3D head. Everything after it is arithmetic — fitting a template, projecting pixels, deciding per point whether a surface faced the camera.

Run the same photograph through twice and the output is byte-identical, down to the last bit. That's what lets the engine debug itself, gate its own output, and refuse a face it can't do well — instead of quietly shipping one that looks wrong.

010203 040506 IntakeThe drawThe fit The buildThe paintInspection refuses badphotos first rough 3D head~5 min, GPU identity asnumbers rebuilds therigged mesh skin texture,25 regions ~40 checkscan refuse DETERMINISTIC — BYTE-IDENTICAL OUTPUT, EVERY RUN
Only stage two invents anything. The other five measure — which is why the same photograph always produces the same character.
DEBUGGABLE

Every defect has a cause

A cause we can find and name — not a model we can only retrain and hope. When one character's jaw looks wrong, we can point to exactly why.

SELF-GATING

It checks its own work

~40 automatic checks compare the result against spec. Any one can refuse the character — the engine stops rather than ships something subtly wrong.

STABLE

Nothing drifts underneath you

Improving the engine never silently changes a character you've already shipped. What you integrated is what stays integrated.

26 JOINTS ONE RIG
§02

Every character.
One skeleton.

We never generate a body. Your photograph supplies identity — expressed as numbers that deform a template we authored, with a fixed topology and a fixed 26-bone skeleton.

Author a motion once and it plays correctly on every character the engine will ever make — which is why we consume NVIDIA Kimodo's text-to-motion output instead of hand-animating each one. A free-form generated mesh can't do this; ours can, because the skeleton never changes.

§03

Where it stands

Built and running today, and what's still in front of us — plainly.

Character from a photographWorkingFourteen characters built and staged, running in Unity
Character from a descriptionWorkingProven end to end with no hand-tuned inputs at all
Real-time on a standalone headsetWorkingWalking, on-device, no PC tether
Automatic refusalWorkingHas refused real subjects and forced real fixes
Hair matched to the subjectDevelopingPlaceholder hair from a shared library today
Full expressive animationDevelopingGait is solved; hands and upper-body detail are next

“Photo-to-avatar platforms have a habit of getting acquired and shut down overnight — developers lose their pipeline with days of notice. Pratima runs as software you operate yourself, on your own infrastructure, so that can't happen to you.”

Building in XR, simulation, or robotics?

Get in touch →