# PrimeComposer: Faster Progressively Combined Diffusion for Image Composition with Attention Steering

arXiv: https://arxiv.org/abs/2403.05053 · HTML: https://mlsanity.com/paper/2403.05053/

## Key facts

- Cited by 20 papers in the map (data through 14 Oct 2026).
- Builds on 43 of its 54 references that are in the map.
- First posted on arXiv 8 Mar 2024.

## Builds on (most cited first, 30 of 43)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [Classifier-Free Diffusion Guidance](https://mlsanity.com/paper/2207.12598/) (2022)
- [DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generation](https://mlsanity.com/paper/2208.12242/) (2022)
- [An Image is Worth One Word: Personalizing Text-to-Image Generation using Textual Inversion](https://mlsanity.com/paper/2208.01618/) (2022)
- [SDEdit: Guided Image Synthesis and Editing with Stochastic Differential Equations](https://mlsanity.com/paper/2108.01073/) (2021)
- [Imagic: Text-Based Real Image Editing with Diffusion Models](https://mlsanity.com/paper/2210.09276/) (2022)
- [Multi-Concept Customization of Text-to-Image Diffusion](https://mlsanity.com/paper/2212.04488/) (2022)
- [DPM-Solver++: Fast Solver for Guided Sampling of Diffusion Probabilistic Models](https://mlsanity.com/paper/2211.01095/) (2022)
- [Blended Diffusion for Text-driven Editing of Natural Images](https://mlsanity.com/paper/2111.14818/) (2021)
- [GLIGEN: Open-Set Grounded Text-to-Image Generation](https://mlsanity.com/paper/2301.07093/) (2023)
- [Attend-and-Excite: Attention-Based Semantic Guidance for Text-to-Image Diffusion Models](https://mlsanity.com/paper/2301.13826/) (2023)
- [Compositional Visual Generation with Composable Diffusion Models](https://mlsanity.com/paper/2206.01714/) (2022)
- [Paint by Example: Exemplar-based Image Editing with Diffusion Models](https://mlsanity.com/paper/2211.13227/) (2022)
- [Blended Latent Diffusion](https://mlsanity.com/paper/2206.02779/) (2022)
- [Training-Free Structured Diffusion Guidance for Compositional Text-to-Image Synthesis](https://mlsanity.com/paper/2212.05032/) (2022)
- [Cut, Paste and Learn: Surprisingly Easy Synthesis for Instance Detection](https://mlsanity.com/paper/1708.01642/) (2017)
- [TF-ICON: Diffusion-Based Training-Free Cross-Domain Image Composition](https://mlsanity.com/paper/2307.12493/) (2023)
- [DoveNet: Deep Image Harmonization via Domain Verification](https://mlsanity.com/paper/1911.13239/) (2019)
- [ST-GAN: Spatial Transformer Generative Adversarial Networks for Image Compositing](https://mlsanity.com/paper/1803.01837/) (2018)
- [Improving the Harmony of the Composite Image by Spatial-Separated Attention Module](https://mlsanity.com/paper/1907.06406/) (2019)
- [GP-GAN: Towards Realistic High-Resolution Image Blending](https://mlsanity.com/paper/1703.07195/) (2017)
- [Making Images Real Again: A Comprehensive Survey on Deep Image Composition](https://mlsanity.com/paper/2106.14490/) (2021)
- [Deep Image Blending](https://mlsanity.com/paper/1910.11495/) (2019)
- [SSH: A Self-Supervised Framework for Image Harmonization](https://mlsanity.com/paper/2108.06805/) (2021)
- [Investigating Prompt Engineering in Diffusion Models](https://mlsanity.com/paper/2211.15462/) (2022)
- [Learning to Generate Synthetic Data via Compositing](https://mlsanity.com/paper/1904.05475/) (2019)

## Built on by (most cited first, 20 of 20)

- [EEdit: Rethinking the Spatial and Temporal Redundancy for Efficient Image Editing](https://mlsanity.com/paper/2503.10270/) (2025)
- [Does FLUX Already Know How to Perform Physically Plausible Image Composition?](https://mlsanity.com/paper/2509.21278/) (2025)
- [UniREditBench: A Unified Reasoning-based Image Editing Benchmark](https://mlsanity.com/paper/2511.01295/) (2025)
- [DreamText: High Fidelity Scene Text Synthesis](https://mlsanity.com/paper/2405.14701/) (2024)
- [Magic Insert: Style-Aware Drag-and-Drop](https://mlsanity.com/paper/2407.02489/) (2024)
- [DreamFuse: Adaptive Image Fusion with Diffusion Transformer](https://mlsanity.com/paper/2504.08291/) (2025)
- [ThinkRL-Edit: Thinking in Reinforcement Learning for Reasoning-Centric Image Editing](https://mlsanity.com/paper/2601.03467/) (2026)
- [MVOC: a training-free multiple video object composition method with diffusion models](https://mlsanity.com/paper/2406.15829/) (2024)
- [MagicFace: Training-free Universal-Style Human Image Customized Synthesis](https://mlsanity.com/paper/2408.07433/) (2024)
- [In-Context Brush: Zero-shot Customized Subject Insertion with Context-Aware Latent Space Manipulation](https://mlsanity.com/paper/2505.20271/) (2025)
- [InsertDiffusion: Identity Preserving Visualization of Objects through a Training-Free Diffusion Architecture](https://mlsanity.com/paper/2407.10592/) (2024)
- [Refine-by-Align: Reference-Guided Artifacts Refinement through Semantic Alignment](https://mlsanity.com/paper/2412.00306/) (2024)
- [AIComposer: Any Style and Content Image Composition via Feature Integration](https://mlsanity.com/paper/2507.20721/) (2025)
- [AnyLogo: Symbiotic Subject-Driven Diffusion System with Gemini Status](https://mlsanity.com/paper/2409.17740/) (2024)
- [Simulate, Refocus and Ensemble: An Attention-Refocusing Scheme for Domain Generalization](https://mlsanity.com/paper/2507.12851/) (2025)
- [Insert In Style: A Zero-Shot Generative Framework for Harmonious Cross-Domain Object Composition](https://mlsanity.com/paper/2511.15197/) (2025)
- [FreeInsert: Personalized Object Insertion with Geometric and Style Control](https://mlsanity.com/paper/2509.20756/) (2025)
- [Reflection Generation for Composite Image Using Diffusion Model](https://mlsanity.com/paper/2604.02168/) (2026)
- [Chameleon: Style-Content Disentangled Framework for Cross-Domain Object Compositing](https://mlsanity.com/paper/2606.01079/) (2026)
- [presto: Efficient, Training-free, and Open-world Object Placement via Imaginary Search](https://mlsanity.com/paper/2608.21543/) (2026)

## Abstract (arXiv)

Image composition involves seamlessly integrating given objects into a specific visual context. Current training-free methods rely on composing attention weights from several samplers to guide the generator. However, since these weights are derived from disparate contexts, their combination leads to coherence confusion and loss of appearance information. These issues worsen with their excessive focus on background generation, even when unnecessary in this task. This not only impedes their swift implementation but also compromises foreground generation quality. Moreover, these methods introduce unwanted artifacts in the transition area. In this paper, we formulate image composition as a subject-based local editing task, solely focusing on foreground generation. At each step, the edited foreground is combined with the noisy background to maintain scene consistency. To address the remaining issues, we propose PrimeComposer, a faster training-free diffuser that composites the images by well-designed attention steering across different noise levels. This steering is predominantly achieved by our Correlation Diffuser, utilizing its self-attention layers at each step. Within these layers, the synthesized subject interacts with both the referenced object and background, capturing intricate details and coherent relationships. This prior information is encoded into the attention weights, which are then integrated into the self-attention layers of the generator to guide the synthesis process. Besides, we introduce a Region-constrained Cross-Attention to confine the impact of specific subject-related tokens to desired regions, addressing the unwanted artifacts shown in the prior method thereby further improving the coherence in the transition area. Our method exhibits the fastest inference efficiency and extensive experiments demonstrate our superiority both qualitatively and quantitatively.