TencentARC/PhotoMaker-V2

PhotoMaker V2 Model Card

Project Page | Paper (ArXiv) | Code

🤗 Gradio demo

Introduction

Users can input one or a few face photos, along with a text prompt, to receive a customized photo or painting within seconds (no training required!). Additionally, this model can be adapted to any base model based on SDXL or used in conjunction with other LoRA modules.

Realistic results

Stylization results

More results can be found in our project page

Model Details

It mainly contains two parts corresponding to two keys in loaded state dict:

1. id_encoder includes finetuned OpenCLIP-ViT-H-14 and a few fuse layers.

2. lora_weights applies to all attention layers in the UNet, and the rank is set to 64.

Usage

You can directly download the model in this repository. You also can download the model in python script:

from huggingface_hub import hf_hub_download
photomaker_ckpt = hf_hub_download(repo_id="TencentARC/PhotoMaker-V2", filename="photomaker-v2.bin", repo_type="model")

Then, please follow the instructions in our GitHub repository.

Limitations

• The model's customization performance degrades on Asian male faces.
• The model still struggles with accurately rendering human hands.

Bias

While the capabilities of image generation models are impressive, they can also reinforce or exacerbate social biases.

Citation

BibTeX:

@inproceedings{li2023photomaker,
  title={PhotoMaker: Customizing Realistic Human Photos via Stacked ID Embedding},
  author={Li, Zhen and Cao, Mingdeng and Wang, Xintao and Qi, Zhongang and Cheng, Ming-Ming and Shan, Ying},
  booktitle={IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
  year={2024}
}