Saved in:
Bibliographic Details
Main Authors: Zhao, Peiang, Li, Han, Jin, Ruiyang, Zhou, S. Kevin
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2311.12342
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909149723361280
author Zhao, Peiang
Li, Han
Jin, Ruiyang
Zhou, S. Kevin
author_facet Zhao, Peiang
Li, Han
Jin, Ruiyang
Zhou, S. Kevin
contents Recent text-to-image diffusion models have reached an unprecedented level in generating high-quality images. However, their exclusive reliance on textual prompts often falls short in precise control of image compositions. In this paper, we propose LoCo, a training-free approach for layout-to-image Synthesis that excels in producing high-quality images aligned with both textual prompts and layout instructions. Specifically, we introduce a Localized Attention Constraint (LAC), leveraging semantic affinity between pixels in self-attention maps to create precise representations of desired objects and effectively ensure the accurate placement of objects in designated regions. We further propose a Padding Token Constraint (PTC) to leverage the semantic information embedded in previously neglected padding tokens, improving the consistency between object appearance and layout instructions. LoCo seamlessly integrates into existing text-to-image and layout-to-image models, enhancing their performance in spatial control and addressing semantic failures observed in prior methods. Extensive experiments showcase the superiority of our approach, surpassing existing state-of-the-art training-free layout-to-image methods both qualitatively and quantitatively across multiple benchmarks.
format Preprint
id arxiv_https___arxiv_org_abs_2311_12342
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle LoCo: Locally Constrained Training-Free Layout-to-Image Synthesis
Zhao, Peiang
Li, Han
Jin, Ruiyang
Zhou, S. Kevin
Computer Vision and Pattern Recognition
Recent text-to-image diffusion models have reached an unprecedented level in generating high-quality images. However, their exclusive reliance on textual prompts often falls short in precise control of image compositions. In this paper, we propose LoCo, a training-free approach for layout-to-image Synthesis that excels in producing high-quality images aligned with both textual prompts and layout instructions. Specifically, we introduce a Localized Attention Constraint (LAC), leveraging semantic affinity between pixels in self-attention maps to create precise representations of desired objects and effectively ensure the accurate placement of objects in designated regions. We further propose a Padding Token Constraint (PTC) to leverage the semantic information embedded in previously neglected padding tokens, improving the consistency between object appearance and layout instructions. LoCo seamlessly integrates into existing text-to-image and layout-to-image models, enhancing their performance in spatial control and addressing semantic failures observed in prior methods. Extensive experiments showcase the superiority of our approach, surpassing existing state-of-the-art training-free layout-to-image methods both qualitatively and quantitatively across multiple benchmarks.
title LoCo: Locally Constrained Training-Free Layout-to-Image Synthesis
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2311.12342