Text2Face: 3D Morphable Faces from Text

William Rowan; Patrik Huber; N. E. Pears; Andrew Keeling

Text2Face: 3D Morphable Faces from Text

William Rowan, Patrik Huber, N. E. Pears, Andrew Keeling

Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

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 imageand-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 multimodal 3DMM image fitting to sketches and sculptures, as well as images.

Original language	English
Title of host publication	International Conference on Learning Representations 2023
Subtitle of host publication	Proceedings
Publisher	IEEE
Number of pages	7
Publication status	Published - 1 May 2023
Event	International Conference on Learning Representations - Kigali, Rwanda Duration: 1 May 2023 → 5 May 2023 Conference number: 11 https://iclr.cc/Conferences/2023

Conference

Conference	International Conference on Learning Representations
Abbreviated title	ICLR
Country/Territory	Rwanda
City	Kigali
Period	1/05/23 → 5/05/23
Internet address	https://iclr.cc/Conferences/2023

Bibliographical note

This is an author-produced version of the published paper. Uploaded in accordance with the University’s Research Publications and Open Access policy.

Keywords

3D morphable model
3D generative face model

Access to Document

text2face_3d_morphable_faces
This is an author-produced version of the published paper. Uploaded in accordance with the University’s Research Publications and Open Access policy.
Accepted author manuscript, 3.81 MBLicence: CC BY

Cite this

@inproceedings{2f9b2c76315c4067bf91d6b8b6f9afe2,

title = "Text2Face: 3D Morphable Faces from Text",

abstract = "We present the first 3D morphable modelling approach, whereby 3D face shapecan be directly and completely defined using a textual prompt. Building on workin multi-modal learning, we extend the FLAME head model to a common imageand-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 policephotofits where the input is already in natural language. It further enables multimodal 3DMM image fitting to sketches and sculptures, as well as images.",

keywords = "3D morphable model, 3D generative face model",

author = "William Rowan and Patrik Huber and Pears, {N. E.} and Andrew Keeling",

note = "This is an author-produced version of the published paper. Uploaded in accordance with the University{\textquoteright}s Research Publications and Open Access policy. ; International Conference on Learning Representations, ICLR ; Conference date: 01-05-2023 Through 05-05-2023",

year = "2023",

month = may,

day = "1",

language = "English",

booktitle = "International Conference on Learning Representations 2023",

publisher = "IEEE",

address = "United States",

url = "https://iclr.cc/Conferences/2023",

}

TY - GEN

T1 - Text2Face: 3D Morphable Faces from Text

AU - Rowan, William

AU - Huber, Patrik

AU - Pears, N. E.

AU - Keeling, Andrew

N1 - Conference code: 11

PY - 2023/5/1

Y1 - 2023/5/1

N2 - We present the first 3D morphable modelling approach, whereby 3D face shapecan be directly and completely defined using a textual prompt. Building on workin multi-modal learning, we extend the FLAME head model to a common imageand-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 policephotofits where the input is already in natural language. It further enables multimodal 3DMM image fitting to sketches and sculptures, as well as images.

AB - We present the first 3D morphable modelling approach, whereby 3D face shapecan be directly and completely defined using a textual prompt. Building on workin multi-modal learning, we extend the FLAME head model to a common imageand-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 policephotofits where the input is already in natural language. It further enables multimodal 3DMM image fitting to sketches and sculptures, as well as images.

KW - 3D morphable model

KW - 3D generative face model

UR - https://openreview.net/pdf?id=7Zyv70nGl_g

M3 - Conference contribution

BT - International Conference on Learning Representations 2023

PB - IEEE

T2 - International Conference on Learning Representations

Y2 - 1 May 2023 through 5 May 2023

ER -

Text2Face: 3D Morphable Faces from Text

Abstract

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Cite this