Controllable Layered Image Generation for Real-World Editing

TL;DR

This work introduces LASAGNA, a unified diffusion-based framework for controllable, real-world image editing via explicit layered representations. It jointly generates a photorealistic background and a transparent foreground with physically grounded visual effects, supporting FG_Gen, BG_Gen, and Text2All conditioning. To enable research and evaluation, the authors present Lasagna-48K, a public dataset with decomposed layers that preserve shadows and reflections, and LasagnaBench, the first public benchmark for layer editing. Empirical results show improved inter-layer coherence, identity preservation, and editing fidelity across multiple tasks, outperforming both general editing models and prior LayerDiffuse-style methods. The work includes extensive ablations and demonstrates practical, creative editing capabilities, with public releases intended to foster open research and standardization in layered image synthesis.

Abstract

Recent image generation models have shown impressive progress, yet they often struggle to yield controllable and consistent results when users attempt to edit specific elements within an existing image. Layered representations enable flexible, user-driven content creation, but existing approaches often fail to produce layers with coherent compositing relationships, and their object layers typically lack realistic visual effects such as shadows and reflections. To overcome these limitations, we propose LASAGNA, a novel, unified framework that generates an image jointly with its composing layers--a photorealistic background and a high-quality transparent foreground with compelling visual effects. Unlike prior work, LASAGNA efficiently learns correct image composition from a wide range of conditioning inputs--text prompts, foreground, background, and location masks--offering greater controllability for real-world applications. To enable this, we introduce LASAGNA-48K, a new dataset composed of clean backgrounds and RGBA foregrounds with physically grounded visual effects. We also propose LASAGNABENCH, the first benchmark for layer editing. We demonstrate that LASAGNA excels in generating highly consistent and coherent results across multiple image layers simultaneously, enabling diverse post-editing applications that accurately preserve identity and visual effects. LASAGNA-48K and LASAGNABENCH will be publicly released to foster open research in the community. The project page is https://rayjryang.github.io/LASAGNA-Page/.

Figures (15)

• Figure 1: Layered generation with Lasagna. (a) Our framework supports three generation modes: background-conditioned foreground generation, foreground-conditioned background generation, and text-to-all layer generation, which flexibly handle different inputs and jointly synthesize coherent, high-quality composites, backgrounds, and transparent foregrounds with realistic visual effects (e.g., shadows and reflections). (b) Generated layers enable direct post-editing into new, coherent scenes.
• Figure 2: Pipeline of Lasagna framework. We formulate the joint generation of composite images, backgrounds, and foregrounds as a flexible, layer-conditional denoising task. This single framework supports multiple workflows, including FG_Gen, BG_Gen, and Text2All. We use a unified input representation with learnable embeddings that distinguish different roles of visual latents (noise, BG, FG, and mask) across tasks, enabling the model to adapt its behavior under various generation settings. This allows a single attention-based model to flexibly process varied combinations of inputs and targets simultaneously.
• Figure 3: Data construction pipeline. Starting with existing datasets, we implement a four-stage data construction pipeline leveraging off-the-shelf models with a custom-trained data curator. This process yields a high-quality dataset as the foundation for subsequent model training.
• Figure 4: Samples of Lasagna-48K and LasagnaBench. Each sample consists of a composite image, a clean background, and a foreground layer with visual effects, along with corresponding captions for all components.
• Figure 5: Layer generation compared with state-of-the-art image generation and editing models. We compare Lasagna with Flux.1 labs2025fluxhuggingface2025flux, Qwen-Image-Edit wu2025qwen, and gpt-image-1[High] openai_gptimage1_2025. (a) Across three distinct generation tasks, Lasagna consistently achieves superior inter-layer coherence and consistency. In contrast, competing models often fail to maintain these properties. (b) Moreover, by generating foregrounds with faithfully preserved visual effects, Lasagna enables diverse post-generation editing operations on individual layers directly—a capability not supported by existing models.