Our mission

We’re building the future of
human-centric commercial photography.

Welcome to the age of neurally rendered hyper-realistic imagery.

The science behind

Based on a deep understanding of 3D face topologies and radiance fields, Smize leverages neural rendering
and facial action units to reenact expressions and emotions onto synthetically generated hyper-realistic human faces.

step 1

Encoding into the latent space
of a GAN

step 2

Editing the semantic parameters
of a 3D morphable model

step 3

Neural rerendering of the target photo

Step 1

Encoding into the latent space of a GAN

step 2

Editing the semantic parameters of a 3D morphable model

step 3

Neural rerendering of the target photo

The team

Matthieu

CSO

Olivier

CEO

Antoine

Research Engineer

Paul

Research Assistant

Benjamin

Research Assistant

Marcel

Research Assistant

Advisory board

Frédéric Rose

Former CEO @ Technicolor

Edmond Boyer

Researcher @ INRIA

David Picard

Researcher @ Ponts

James O'Brien

Researcher @ Berkeley

Dmitriy Smirnov

Researcher @ MIT

James O'Brien

Researcher @ Berkeley

Dmitriy Smirnov

Researcher @ MIT

Join the beta

Get early access to Smize and start turning failed shots into your best shots.

Request Access