Every expression your character needs is one photograph away.

ImageRig reads a single photograph and writes the full ARKit expression straight onto your character's shape keys — inside Blender, completely offline, in under a second. Need a snarl for the close-up, a raised brow for the story beat, a grin for the poster? The photo is the direction; ImageRig is the performance.

And yes — the name sounds familiar on purpose: ImageRig is the little brother of AudioRig, our Super Hive Market add-on that makes the same ARKit characters perform, baking full facial animation and lip-sync from an audio file. ImageRig poses the frame. AudioRig plays the scene.

Same rig. Same scene. One photograph in between.

Before Your character, resting
Source Any photograph you direct with
After Generated + applied, no sculpting
Take 01 · The grin
Neutral render of the character, take 1
Source photograph, take 1
Generated expression applied to the rig, take 1
Take 02 · The fear
Neutral render of the character, take 2
Source photograph, take 2
Generated expression applied to the rig, take 2
Take 03 · The wink
Neutral render of the character, take 3
Source photograph, take 3
Generated expression applied to the rig, take 3

From still to performance in three moves.

No rigging sessions, no shape-key sculpting, no export round-trips. ImageRig lives in your sidebar.

Pick the photograph

Any photo of a face making the expression you need — the director's reference, a frame of an actor's audition, your own camera roll. The closer to frontal, the better.

Picking the source photograph

Press Generate + Apply

ImageRig reads the 52 ARKit coefficients from the photo, cleans them with the quality pipeline, and writes them straight onto your rig's shape keys — eyes, teeth and tongue included.

The Generate + Apply button in the sidebar

Direct the performance

Silence micro-noise with the threshold, strip resting bias with a neutral capture, or pick a calibration preset for stylized rigs. Every result stays editable — shape keys are just shape keys.

Fine-tuning the result

Built for production, not demos.

Completely offline & private.

The face engine and every component ship inside the extension. No account, no uploads, no telemetry — your photos never leave Blender.

Your rig already speaks ARKit.

MetaHuman exports, Rigify face rigs, marketplace head rigs: if the shape keys use the ARKit 52 naming, ImageRig finds them and fills them.

Under a second per expression.

Asynchronous pipeline: Blender stays responsive, a progress bar tracks the run, and the expression lands while you watch the viewport.

READ THE FULL DOCUMENTATION

One face. Endless direction.

Every card below is one Generate + Apply away — one face, six directions.

Joy expression generated from a photograph
Joy

Wide grin, cheeks up — the poster-frame smile, straight from a selfie.
Fear expression generated from a photograph
Fear

Brows pinned high, eyes wide — the beat before the turn.
Surprise expression generated from a photograph
Surprise

Jaw dropped, brows high — the reaction shot you were missing.
Disgust expression generated from a photograph
Disgust

Nose sneer and lip raise — the micro-expression rigs never get right.
Half-side sneer expression generated from a photograph
Sneer

Half snarl, one side — asymmetric direction from an asymmetric photo.
Winking expression generated from a photograph
Wink

One eye closed, smile loading — personality in a single frame.
TIME-LAPSE · NEUTRAL → EXPRESSION
Watch a take come together

A time-lapse of the full loop: pick the photo, press Generate + Apply, watch the face move.

Put your next expression
one click away.

ImageRig is available on Super Hive Market. And when your character needs to talk — its older brother AudioRig is already there.

GET IMAGERIG ON SUPER HIVE MARKET MEET AUDIORIG