---
license: openrail++
library_name: diffusers
tags:
- text-to-image
- stable-diffusion
---

# Conceptrol: Concept Control of Zero-shot Personalized Image Generation

## Model Card

This model implements Conceptrol, a training-free method that boosts zero-shot personalized image generation across Stable Diffusion, SDXL, and FLUX. It works without additional training, data, or models.

<p align="center">
  <img src="demo/teaser.png">
</p>

[Conceptrol: Concept Control of Zero-shot Personalized Image Generation](https://huggingface.co/papers/2503.06568)

**Abstract:**

Personalized image generation with text-to-image diffusion models generates unseen images based on reference image content. Zero-shot adapter methods such as IP-Adapter and OminiControl are especially interesting because they do not require test-time fine-tuning. However, they struggle to balance preserving personalized content and adherence to the text prompt.  We identify a critical design flaw resulting in this performance gap: current adapters inadequately integrate personalization images with the textual descriptions. The generated images, therefore, replicate the personalized content rather than adhere to the text prompt instructions. Yet the base text-to-image has strong conceptual understanding capabilities that can be leveraged.

We propose Conceptrol, a simple yet effective framework that enhances zero-shot adapters without adding computational overhead. Conceptrol constrains the attention of visual specification with a textual concept mask that improves subject-driven generation capabilities. It achieves as much as 89% improvement on personalization benchmarks over the vanilla IP-Adapter and can even outperform fine-tuning approaches such as Dreambooth LoRA.

## Quick Start

#### 1. Environment Setup

``` bash
conda create -n conceptrol python=3.10
conda activate conceptrol
pip install -r requirements.txt
```

#### 2. Go to `demo_sd.ipynb` / `demo_sdxl.ipynb` / `demo_flux.py` for fun!

## Local Setup using Gradio

#### 1. Start Gradio Interface
``` bash
pip install gradio
gradio gradio_src/app.py
```

#### 2. Use the GUI!

## Supporting Models

| Model Name            |  Link                                             |
|-----------------------|-------------------------------------------------------------|
| Stable Diffusion 1.5  | [stable-diffusion-v1-5/stable-diffusion-v1-5](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5)   |
| Realistic Vision V5.1 | [SG161222/Realistic_Vision_V5.1_noVAE](https://huggingface.co/SG161222/Realistic_Vision_V5.1_noVAE) |
| Stable Diffusion XL-1024   | [stabilityai/stable-diffusion-xl-base-1.0](https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0) |
| Animagine XL v4.0 |   [cagliostrolab/animagine-xl-4.0](https://huggingface.co/cagliostrolab/animagine-xl-4.0)|
| Realistic Vision XL V5.0 | [SG161222/RealVisXL_V5.0](https://huggingface.co/SG161222/RealVisXL_V5.0) |
| FLUX-schnell | [black-forest-labs/FLUX.1-schnell](https://huggingface.co/black-forest-labs/FLUX.1-schnell) |

| Adapter Name            |  Link                                             |
|-----------------------|-------------------------------------------------------------|
| IP-Adapter  | [h94/IP-Adapter](https://huggingface.co/h94/IP-Adapter/tree/main)  |
| OminiControl | [Yuanshi/OminiControl](https://huggingface.co/Yuanshi/OminiControl) |


## Source Code

https://github.com/QY-H00/Conceptrol

## Citation

``` bibtex
@article{he2025conceptrol,
  title={Conceptrol: Concept Control of Zero-shot Personalized Image Generation},
  author={Qiyuan He and Angela Yao},
  journal={arXiv preprint arXiv:2503.06568},
  year={2025}
}
```

## Acknowledgement

We thank the following repositories for their great work: 

[diffusers](https://github.com/huggingface/diffusers), 
[transformers](https://github.com/huggingface/transformers), 
[IP-Adapter](https://github.com/tencent-ailab/IP-Adapter), 
[OminiControl](https://github.com/Yuanshi9815/OminiControl)