MagicQuillV2: Precise and Interactive Image Editing with Layered Visual Cues

TL;DR

MagicQuillV2 introduces a layered visual cue system for precise, interactive image editing, combining diffusion models' semantic capabilities with granular control akin to traditional graphics software.

Contribution

It presents a novel layered composition paradigm, including a new data pipeline, control module, and spatial editing branch for improved user control in generative image editing.

Findings

Effective disentanglement of user intentions into visual layers

Enhanced control over content, position, shape, and color in generated images

Validated through extensive experiments demonstrating improved editing precision

Abstract

We propose MagicQuill V2, a novel system that introduces a \textbf{layered composition} paradigm to generative image editing, bridging the gap between the semantic power of diffusion models and the granular control of traditional graphics software. While diffusion transformers excel at holistic generation, their use of singular, monolithic prompts fails to disentangle distinct user intentions for content, position, and appearance. To overcome this, our method deconstructs creative intent into a stack of controllable visual cues: a content layer for what to create, a spatial layer for where to place it, a structural layer for how it is shaped, and a color layer for its palette. Our technical contributions include a specialized data generation pipeline for context-aware content integration, a unified control module to process all visual cues, and a fine-tuned spatial branch for precise local editing, including object removal. Extensive experiments validate that this layered approach effectively resolves the user intention gap, granting creators direct, intuitive control over the generative process.