LimeCross: Context-Conditioned Layered Image Editing with Structural Consistency
Ryugo Morita, Stanislav Frolov, Brian Bernhard Moser, Ko Watanabe, Riku Takahashi, Issey Sukeda, Andreas Dengel

TL;DR
LimeCross is a training-free framework for context-aware layered image editing that preserves structural consistency and layer integrity during text-guided modifications.
Contribution
It introduces a novel attention-based method for editing RGBA layers with context-awareness, avoiding layer collapse and background leakage.
Findings
LimeCross outperforms existing baselines in layer purity.
It maintains high composite realism in edited images.
The approach effectively preserves layer structure and transparency.
Abstract
Layered image assets are widely used in real-world creative workflows, enabling non-destructive iteration and flexible re-composition. Recent advances in layered image generation and decomposition synthesize or recover layered representations, yet controllable editing of layered images remains challenging. Manual editing requires careful coordination across layers to maintain consistent illumination and contact, while AI-based pipelines collapse layers into a flattened image for editing, then decompose them again, introducing background-to-foreground leakage and unstable transparency. To address these limitations, we propose LimeCross, a training-free context-conditioned layered image editing framework that edits user-selected RGBA layers according to text while keeping the remaining layers unchanged. It leverages contextual cues from other layers using a bi-stream attention mechanism…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
