DeepAI AI Chat
Log In Sign Up

Learned Spatial Representations for Few-shot Talking-Head Synthesis

by   Moustafa Meshry, et al.

We propose a novel approach for few-shot talking-head synthesis. While recent works in neural talking heads have produced promising results, they can still produce images that do not preserve the identity of the subject in source images. We posit this is a result of the entangled representation of each subject in a single latent code that models 3D shape information, identity cues, colors, lighting and even background details. In contrast, we propose to factorize the representation of a subject into its spatial and style components. Our method generates a target frame in two steps. First, it predicts a dense spatial layout for the target image. Second, an image generator utilizes the predicted layout for spatial denormalization and synthesizes the target frame. We experimentally show that this disentangled representation leads to a significant improvement over previous methods, both quantitatively and qualitatively.


page 5

page 7

page 8

page 14

page 15

page 16

page 17

page 18


StyleMask: Disentangling the Style Space of StyleGAN2 for Neural Face Reenactment

In this paper we address the problem of neural face reenactment, where, ...

Controllable Radiance Fields for Dynamic Face Synthesis

Recent work on 3D-aware image synthesis has achieved compelling results ...

HeadGAN: Video-and-Audio-Driven Talking Head Synthesis

Recent attempts to solve the problem of talking head synthesis using a s...

Few-Shot Adversarial Learning of Realistic Neural Talking Head Models

Several recent works have shown how highly realistic human head images c...

One-Shot Identity-Preserving Portrait Reenactment

We present a deep learning-based framework for portrait reenactment from...

Plug-in Factorization for Latent Representation Disentanglement

In this work, we propose a Factorized Disentangler-Entangler Network (FD...