Text2Face: 3D Morphable Faces From Text

Will Rowan, Patrik Huber, Nick Pears, Andrew Keeling

Keywords: 3D Face Generation, Multi-Modal Face Fitting, 3D Morphable Models

TL;DR: We present the first method to generate fully parameterised 3D face shape, including identity, from natural language descriptors.

Abstract: We present the first 3D morphable modelling approach, whereby 3D face shape can be directly and completely defined using a textual prompt. Building on work in multi-modal learning, we extend the FLAME head model to a common image-and-text latent space. This allows for direct 3D Morphable Model (3DMM) parameter generation and therefore shape manipulation from textual descriptions. Our method, Text2Face, has many applications; for example: generating police photofits where the input is already in natural language. It further enables multi- modal 3DMM image fitting to sketches and sculptures, as well as images.