Saved in:
Bibliographic Details
Main Authors: Li, Leheng, Qiu, Weichao, Yan, Xu, He, Jing, Zhou, Kaiqiang, Cai, Yingjie, Lian, Qing, Liu, Bingbing, Chen, Ying-Cong
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2410.04932
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914966573940736
author Li, Leheng
Qiu, Weichao
Yan, Xu
He, Jing
Zhou, Kaiqiang
Cai, Yingjie
Lian, Qing
Liu, Bingbing
Chen, Ying-Cong
author_facet Li, Leheng
Qiu, Weichao
Yan, Xu
He, Jing
Zhou, Kaiqiang
Cai, Yingjie
Lian, Qing
Liu, Bingbing
Chen, Ying-Cong
contents We present OmniBooth, an image generation framework that enables spatial control with instance-level multi-modal customization. For all instances, the multimodal instruction can be described through text prompts or image references. Given a set of user-defined masks and associated text or image guidance, our objective is to generate an image, where multiple objects are positioned at specified coordinates and their attributes are precisely aligned with the corresponding guidance. This approach significantly expands the scope of text-to-image generation, and elevates it to a more versatile and practical dimension in controllability. In this paper, our core contribution lies in the proposed latent control signals, a high-dimensional spatial feature that provides a unified representation to integrate the spatial, textual, and image conditions seamlessly. The text condition extends ControlNet to provide instance-level open-vocabulary generation. The image condition further enables fine-grained control with personalized identity. In practice, our method empowers users with more flexibility in controllable generation, as users can choose multi-modal conditions from text or images as needed. Furthermore, thorough experiments demonstrate our enhanced performance in image synthesis fidelity and alignment across different tasks and datasets. Project page: https://len-li.github.io/omnibooth-web/
format Preprint
id arxiv_https___arxiv_org_abs_2410_04932
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle OmniBooth: Learning Latent Control for Image Synthesis with Multi-modal Instruction
Li, Leheng
Qiu, Weichao
Yan, Xu
He, Jing
Zhou, Kaiqiang
Cai, Yingjie
Lian, Qing
Liu, Bingbing
Chen, Ying-Cong
Computer Vision and Pattern Recognition
We present OmniBooth, an image generation framework that enables spatial control with instance-level multi-modal customization. For all instances, the multimodal instruction can be described through text prompts or image references. Given a set of user-defined masks and associated text or image guidance, our objective is to generate an image, where multiple objects are positioned at specified coordinates and their attributes are precisely aligned with the corresponding guidance. This approach significantly expands the scope of text-to-image generation, and elevates it to a more versatile and practical dimension in controllability. In this paper, our core contribution lies in the proposed latent control signals, a high-dimensional spatial feature that provides a unified representation to integrate the spatial, textual, and image conditions seamlessly. The text condition extends ControlNet to provide instance-level open-vocabulary generation. The image condition further enables fine-grained control with personalized identity. In practice, our method empowers users with more flexibility in controllable generation, as users can choose multi-modal conditions from text or images as needed. Furthermore, thorough experiments demonstrate our enhanced performance in image synthesis fidelity and alignment across different tasks and datasets. Project page: https://len-li.github.io/omnibooth-web/
title OmniBooth: Learning Latent Control for Image Synthesis with Multi-modal Instruction
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2410.04932