Insert In Style: A Zero-Shot Generative Framework for Harmonious Cross-Domain Object Composition

TL;DR

Insert In Style is a novel zero-shot generative framework for cross-domain object composition that achieves high fidelity and style harmony without online finetuning, using a multi-stage training protocol and specialized architecture.

Contribution

The paper introduces a unified zero-shot framework with disentangled representations and a masked-attention architecture for harmonious cross-domain object composition.

Findings

Outperforms existing methods on identity and style metrics.

Achieves state-of-the-art results in high-fidelity stylized composition.

Validated by user studies confirming improved performance.

Abstract

Reference-based object composition involves integrating foreground reference image with background scene to produce harmonious fused image. This task becomes particularly challenging in cross-domain scenarios, where models must balance preserving the reference object's identity while harmonizing them to match stylized environments. This under-explored problem is currently split between practical "blenders" that lack generative fidelity and "generators" that require impractical, per-subject online finetuning. In this work, we introduce Insert In Style, the first zero-shot generative framework that is both practical and high-fidelity. Our core contribution is a unified framework with two key innovations: (i) a novel multi-stage training protocol that disentangles representations for identity, style, and composition, and (ii) a specialized masked-attention architecture that surgically enforces this disentanglement during generation (iii) A prior preservation objective that keeps learned identity and style priors intact. By design, this approach mitigates concept interference typical in unified-attention architectures while ensuring robust generalization across diverse references and styles. Our framework is trained on a new 115k sample dataset, curated from a novel data pipeline. This pipeline couples large-scale generation with a rigorous, iterative human-in-the-loop filtering process to ensure both high-fidelity semantic identity and style coherence. Unlike prior work, our model is truly zero-shot and requires no text prompts. We also introduce a new public benchmark for stylized composition. We demonstrate state-of-the-art performance, significantly outperforming existing methods on both identity and style metrics, a result strongly corroborated by user studies.