Saved in:
Bibliographic Details
Main Authors: Han, Yue, Zhu, Junwei, He, Keke, Chen, Xu, Ge, Yanhao, Li, Wei, Li, Xiangtai, Zhang, Jiangning, Wang, Chengjie, Liu, Yong
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2405.12970
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911948671549440
author Han, Yue
Zhu, Junwei
He, Keke
Chen, Xu
Ge, Yanhao
Li, Wei
Li, Xiangtai
Zhang, Jiangning
Wang, Chengjie
Liu, Yong
author_facet Han, Yue
Zhu, Junwei
He, Keke
Chen, Xu
Ge, Yanhao
Li, Wei
Li, Xiangtai
Zhang, Jiangning
Wang, Chengjie
Liu, Yong
contents Current face reenactment and swapping methods mainly rely on GAN frameworks, but recent focus has shifted to pre-trained diffusion models for their superior generation capabilities. However, training these models is resource-intensive, and the results have not yet achieved satisfactory performance levels. To address this issue, we introduce Face-Adapter, an efficient and effective adapter designed for high-precision and high-fidelity face editing for pre-trained diffusion models. We observe that both face reenactment/swapping tasks essentially involve combinations of target structure, ID and attribute. We aim to sufficiently decouple the control of these factors to achieve both tasks in one model. Specifically, our method contains: 1) A Spatial Condition Generator that provides precise landmarks and background; 2) A Plug-and-play Identity Encoder that transfers face embeddings to the text space by a transformer decoder. 3) An Attribute Controller that integrates spatial conditions and detailed attributes. Face-Adapter achieves comparable or even superior performance in terms of motion control precision, ID retention capability, and generation quality compared to fully fine-tuned face reenactment/swapping models. Additionally, Face-Adapter seamlessly integrates with various StableDiffusion models.
format Preprint
id arxiv_https___arxiv_org_abs_2405_12970
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Face Adapter for Pre-Trained Diffusion Models with Fine-Grained ID and Attribute Control
Han, Yue
Zhu, Junwei
He, Keke
Chen, Xu
Ge, Yanhao
Li, Wei
Li, Xiangtai
Zhang, Jiangning
Wang, Chengjie
Liu, Yong
Computer Vision and Pattern Recognition
Current face reenactment and swapping methods mainly rely on GAN frameworks, but recent focus has shifted to pre-trained diffusion models for their superior generation capabilities. However, training these models is resource-intensive, and the results have not yet achieved satisfactory performance levels. To address this issue, we introduce Face-Adapter, an efficient and effective adapter designed for high-precision and high-fidelity face editing for pre-trained diffusion models. We observe that both face reenactment/swapping tasks essentially involve combinations of target structure, ID and attribute. We aim to sufficiently decouple the control of these factors to achieve both tasks in one model. Specifically, our method contains: 1) A Spatial Condition Generator that provides precise landmarks and background; 2) A Plug-and-play Identity Encoder that transfers face embeddings to the text space by a transformer decoder. 3) An Attribute Controller that integrates spatial conditions and detailed attributes. Face-Adapter achieves comparable or even superior performance in terms of motion control precision, ID retention capability, and generation quality compared to fully fine-tuned face reenactment/swapping models. Additionally, Face-Adapter seamlessly integrates with various StableDiffusion models.
title Face Adapter for Pre-Trained Diffusion Models with Fine-Grained ID and Attribute Control
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2405.12970