Saved in:
Bibliographic Details
Main Authors: Lungu-Stan, Vlad-Constantin, Mironica, Ionut, Georgescu, Mariana-Iuliana
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2603.17965
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915872943112192
author Lungu-Stan, Vlad-Constantin
Mironica, Ionut
Georgescu, Mariana-Iuliana
author_facet Lungu-Stan, Vlad-Constantin
Mironica, Ionut
Georgescu, Mariana-Iuliana
contents Media design layer generation enables the creation of fully editable, layered design documents such as posters, flyers, and logos using only natural language prompts. Existing methods either restrict outputs to a fixed number of layers or require each layer to contain only spatially continuous regions, causing the layer count to scale linearly with design complexity. We propose LaDe (Layered Media Design), a latent diffusion framework that generates a flexible number of semantically meaningful layers. LaDe combines three components: an LLM-based prompt expander that transforms a short user intent into structured per-layer descriptions that guide the generation, a Latent Diffusion Transformer with a 4D RoPE positional encoding mechanism that jointly generates the full media design and its constituent RGBA layers, and an RGBA VAE that decodes each layer with full alpha-channel support. By conditioning on layer samples during training, our unified framework supports three tasks: text-to-image generation, text-to-layers media design generation, and media design decomposition. We compare LaDe to Qwen-Image-Layered on text-to-layers and image-to-layers tasks on the Crello test set. LaDe outperforms Qwen-Image-Layered in text-to-layers generation by improving text-to-layer alignment, as validated by two VLM-as-a-judge evaluators (GPT-4o mini and Qwen3-VL).
format Preprint
id arxiv_https___arxiv_org_abs_2603_17965
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle LaDe: Unified Multi-Layered Graphic Media Generation and Decomposition
Lungu-Stan, Vlad-Constantin
Mironica, Ionut
Georgescu, Mariana-Iuliana
Computer Vision and Pattern Recognition
Media design layer generation enables the creation of fully editable, layered design documents such as posters, flyers, and logos using only natural language prompts. Existing methods either restrict outputs to a fixed number of layers or require each layer to contain only spatially continuous regions, causing the layer count to scale linearly with design complexity. We propose LaDe (Layered Media Design), a latent diffusion framework that generates a flexible number of semantically meaningful layers. LaDe combines three components: an LLM-based prompt expander that transforms a short user intent into structured per-layer descriptions that guide the generation, a Latent Diffusion Transformer with a 4D RoPE positional encoding mechanism that jointly generates the full media design and its constituent RGBA layers, and an RGBA VAE that decodes each layer with full alpha-channel support. By conditioning on layer samples during training, our unified framework supports three tasks: text-to-image generation, text-to-layers media design generation, and media design decomposition. We compare LaDe to Qwen-Image-Layered on text-to-layers and image-to-layers tasks on the Crello test set. LaDe outperforms Qwen-Image-Layered in text-to-layers generation by improving text-to-layer alignment, as validated by two VLM-as-a-judge evaluators (GPT-4o mini and Qwen3-VL).
title LaDe: Unified Multi-Layered Graphic Media Generation and Decomposition
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2603.17965