MICo-150K: A Comprehensive Dataset Advancing Multi-Image Composition

TL;DR

This paper introduces MICo-150K, a large-scale high-quality dataset for multi-image composition, along with benchmarks and a new evaluation metric, to advance research in controllable image generation from multiple references.

Contribution

The authors curated MICo-150K, a comprehensive dataset with diverse MICo prompts, and developed MICo-Bench and a new metric, enabling better evaluation and training of MICo models.

Findings

MICo-150K improves model capabilities in multi-image composition.

Fine-tuned models on MICo-150K outperform previous methods.

Qwen-MICo supports arbitrary multi-image inputs and matches existing models in 3-image composition.

Abstract

In controllable image generation, synthesizing coherent and consistent images from multiple reference inputs, i.e., Multi-Image Composition (MICo), remains a challenging problem, partly hindered by the lack of high-quality training data. To bridge this gap, we conduct a systematic study of MICo, categorizing it into 7 representative tasks and curate a large-scale collection of high-quality source images and construct diverse MICo prompts. Leveraging powerful proprietary models, we synthesize a rich amount of balanced composite images, followed by human-in-the-loop filtering and refinement, resulting in MICo-150K, a comprehensive dataset for MICo with identity consistency. We further build a Decomposition-and-Recomposition (De&Re) subset, where 11K real-world complex images are decomposed into components and recomposed, enabling both real and synthetic compositions. To enable comprehensive evaluation, we construct MICo-Bench with 100 cases per task and 300 challenging De&Re cases, and further introduce a new metric, Weighted-Ref-VIEScore, specifically tailored for MICo evaluation. Finally, we fine-tune multiple models on MICo-150K and evaluate them on MICo-Bench. The results show that MICo-150K effectively equips models without MICo capability and further enhances those with existing skills. Notably, our baseline model, Qwen-MICo, fine-tuned from Qwen-Image-Edit, matches Qwen-Image-2509 in 3-image composition while supporting arbitrary multi-image inputs beyond the latter's limitation. Our dataset, benchmark, and baseline collectively offer valuable resources for further research on Multi-Image Composition.