LayerFusion: Harmonized Multi-Layer Text-to-Image Generation with Generative Priors

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Dalva, Yusuf, Li, Yijun, Liu, Qing, Zhao, Nanxuan, Zhang, Jianming, Lin, Zhe, Yanardag, Pinar
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909417637675008
author Dalva, Yusuf
Li, Yijun
Liu, Qing
Zhao, Nanxuan
Zhang, Jianming
Lin, Zhe
Yanardag, Pinar
author_facet Dalva, Yusuf
Li, Yijun
Liu, Qing
Zhao, Nanxuan
Zhang, Jianming
Lin, Zhe
Yanardag, Pinar
contents Large-scale diffusion models have achieved remarkable success in generating high-quality images from textual descriptions, gaining popularity across various applications. However, the generation of layered content, such as transparent images with foreground and background layers, remains an under-explored area. Layered content generation is crucial for creative workflows in fields like graphic design, animation, and digital art, where layer-based approaches are fundamental for flexible editing and composition. In this paper, we propose a novel image generation pipeline based on Latent Diffusion Models (LDMs) that generates images with two layers: a foreground layer (RGBA) with transparency information and a background layer (RGB). Unlike existing methods that generate these layers sequentially, our approach introduces a harmonized generation mechanism that enables dynamic interactions between the layers for more coherent outputs. We demonstrate the effectiveness of our method through extensive qualitative and quantitative experiments, showing significant improvements in visual coherence, image quality, and layer consistency compared to baseline methods.
format Preprint
id arxiv_https___arxiv_org_abs_2412_04460
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle LayerFusion: Harmonized Multi-Layer Text-to-Image Generation with Generative Priors
Dalva, Yusuf
Li, Yijun
Liu, Qing
Zhao, Nanxuan
Zhang, Jianming
Lin, Zhe
Yanardag, Pinar
Computer Vision and Pattern Recognition
Large-scale diffusion models have achieved remarkable success in generating high-quality images from textual descriptions, gaining popularity across various applications. However, the generation of layered content, such as transparent images with foreground and background layers, remains an under-explored area. Layered content generation is crucial for creative workflows in fields like graphic design, animation, and digital art, where layer-based approaches are fundamental for flexible editing and composition. In this paper, we propose a novel image generation pipeline based on Latent Diffusion Models (LDMs) that generates images with two layers: a foreground layer (RGBA) with transparency information and a background layer (RGB). Unlike existing methods that generate these layers sequentially, our approach introduces a harmonized generation mechanism that enables dynamic interactions between the layers for more coherent outputs. We demonstrate the effectiveness of our method through extensive qualitative and quantitative experiments, showing significant improvements in visual coherence, image quality, and layer consistency compared to baseline methods.
title LayerFusion: Harmonized Multi-Layer Text-to-Image Generation with Generative Priors
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2412.04460