base_model:
- black-forest-labs/FLUX.1-Fill-dev
language:
- en
license: other
license_name: community-license-agreement
license_link: LICENSE
pipeline_tag: image-to-image
tags:
- image customization
IC-Custom: Diverse Image Customization via In-Context Learning
Abstract
Image customization, a crucial technique for industrial media production, aims to generate content that is consistent with reference images. However, current approaches conventionally separate image customization into position-aware and position-free customization paradigms and lack a universal framework for diverse customization, limiting their applications across various scenarios. To overcome these limitations, we propose IC-Custom, a unified framework that seamlessly integrates position-aware and position-free image customization through in-context learning. IC-Custom concatenates reference images with target images to a polyptych, leveraging DiT's multi-modal attention mechanism for fine-grained token-level interactions. We introduce the In-context Multi-Modal Attention (ICMA) mechanism with learnable task-oriented register tokens and boundary-aware positional embeddings to enable the model to correctly handle different task types and distinguish various inputs in polyptych configurations. To bridge the data gap, we carefully curated a high-quality dataset of 12k identity-consistent samples with 8k from real-world sources and 4k from high-quality synthetic data, avoiding the overly glossy and over-saturated synthetic appearance. IC-Custom supports various industrial applications, including try-on, accessory placement, furniture arrangement, and creative IP customization. Extensive evaluations on our proposed ProductBench and the publicly available DreamBench demonstrate that IC-Custom significantly outperforms community workflows, closed-source models, and state-of-the-art open-source approaches. IC-Custom achieves approximately 73% higher human preference across identity consistency, harmonicity, and text alignment metrics, while training only 0.4% of the original model parameters.
IC-Custom is designed for diverse image customization scenarios, including:
Position-aware: Input a reference image, target background, and specify the customization location (via segmentation or drawing)
Examples: Product placement, virtual try-on.Position-free: Input a reference image and a target description to generate a new image with the reference image's ID
Examples: IP customization, character creation.
Citation
@article{li2025iccustom,
title={IC-Custom: Diverse Image Customization via In-Context Learning},
author={Li, Yaowei and Zhu, Yu and Wu, Xu and Liu, Bo and Li, Jia and Lu, Yong and Zhang, Song and Luo, Yujun},
journal={arXiv preprint arXiv:2507.01926},
year={2025},
url={https://arxiv.org/abs/2507.01926}
}