Arc2Face: A Foundation Model for ID-Consistent Human Faces

Foivos Paraperas Papantoniou¹ Alexandros Lattas¹ Stylianos Moschoglou¹

Jiankang Deng¹ Bernhard Kainz^1,2 Stefanos Zafeiriou¹

¹Imperial College London, UK
²FAU Erlangen-Nürnberg, Germany

This is the official implementation of Arc2Face, an ID-conditioned face model:

✅ that generates high-quality images of any subject given only its ArcFace embedding, within a few seconds
✅ trained on the large-scale WebFace42M dataset offers superior ID similarity compared to existing models
✅ built on top of Stable Diffusion, can be extended to different input modalities, e.g. with ControlNet

News/Updates

[2024/08/16] 🔥 Accepted to ECCV24 as an oral!
[2024/08/06] 🔥 ComfyUI support available at caleboleary/ComfyUI-Arc2Face!
[2024/04/12] 🔥 We add LCM-LoRA support for even faster inference (check the details below).
[2024/04/11] 🔥 We release the training dataset on HuggingFace Datasets.
[2024/03/31] 🔥 We release our demo for pose control using Arc2Face + ControlNet (see instructions below).
[2024/03/28] 🔥 We release our Gradio demo on HuggingFace Spaces (thanks to the HF team for their free GPU support)!
[2024/03/14] 🔥 We release Arc2Face.

Installation

conda create -n arc2face python=3.10
conda activate arc2face

# Install requirements
pip install -r requirements.txt

Download Models

The models can be downloaded manually from HuggingFace or using python:

from huggingface_hub import hf_hub_download

hf_hub_download(repo_id="FoivosPar/Arc2Face", filename="arc2face/config.json", local_dir="./models")
hf_hub_download(repo_id="FoivosPar/Arc2Face", filename="arc2face/diffusion_pytorch_model.safetensors", local_dir="./models")
hf_hub_download(repo_id="FoivosPar/Arc2Face", filename="encoder/config.json", local_dir="./models")
hf_hub_download(repo_id="FoivosPar/Arc2Face", filename="encoder/pytorch_model.bin", local_dir="./models")

For face detection and ID-embedding extraction, manually download the antelopev2 package (direct link) and place the checkpoints under models/antelopev2.
We use an ArcFace recognition model trained on WebFace42M. Download arcface.onnx from HuggingFace and put it in models/antelopev2 or using python:

hf_hub_download(repo_id="FoivosPar/Arc2Face", filename="arcface.onnx", local_dir="./models/antelopev2")

Then delete glintr100.onnx (the default backbone from insightface).

The models folder structure should finally be:

  . ── models ──┌── antelopev2
                ├── arc2face
                └── encoder

Usage

Load pipeline using diffusers:

from diffusers import (
    StableDiffusionPipeline,
    UNet2DConditionModel,
    DPMSolverMultistepScheduler,
)

from arc2face import CLIPTextModelWrapper, project_face_embs

import torch
from insightface.app import FaceAnalysis
from PIL import Image
import numpy as np

# Arc2Face is built upon SD1.5
# The repo below can be used instead of the now deprecated 'runwayml/stable-diffusion-v1-5'
base_model = 'stable-diffusion-v1-5/stable-diffusion-v1-5'

encoder = CLIPTextModelWrapper.from_pretrained(
    'models', subfolder="encoder", torch_dtype=torch.float16
)

unet = UNet2DConditionModel.from_pretrained(
    'models', subfolder="arc2face", torch_dtype=torch.float16
)

pipeline = StableDiffusionPipeline.from_pretrained(
        base_model,
        text_encoder=encoder,
        unet=unet,
        torch_dtype=torch.float16,
        safety_checker=None
    )

You can use any SD-compatible schedulers and steps, just like with Stable Diffusion. By default, we use DPMSolverMultistepScheduler with 25 steps, which produces very good results in just a few seconds.

pipeline.scheduler = DPMSolverMultistepScheduler.from_config(pipeline.scheduler.config)
pipeline = pipeline.to('cuda')

Pick an image and extract the ID-embedding:

app = FaceAnalysis(name='antelopev2', root='./', providers=['CUDAExecutionProvider', 'CPUExecutionProvider'])
app.prepare(ctx_id=0, det_size=(640, 640))

img = np.array(Image.open('assets/examples/joacquin.png'))[:,:,::-1]

faces = app.get(img)
faces = sorted(faces, key=lambda x:(x['bbox'][2]-x['bbox'][0])*(x['bbox'][3]-x['bbox'][1]))[-1]  # select largest face (if more than one detected)
id_emb = torch.tensor(faces['embedding'], dtype=torch.float16)[None].cuda()
id_emb = id_emb/torch.norm(id_emb, dim=1, keepdim=True)   # normalize embedding
id_emb = project_face_embs(pipeline, id_emb)    # pass through the encoder

Generate images:

num_images = 4
images = pipeline(prompt_embeds=id_emb, num_inference_steps=25, guidance_scale=3.0, num_images_per_prompt=num_images).images

LCM-LoRA acceleration

LCM-LoRA allows you to reduce the sampling steps to as few as 2-4 for super-fast inference. Just plug in the pre-trained distillation adapter for SD v1.5 and switch to LCMScheduler:

from diffusers import LCMScheduler

pipeline.load_lora_weights("latent-consistency/lcm-lora-sdv1-5")
pipeline.scheduler = LCMScheduler.from_config(pipeline.scheduler.config)

Then, you can sample with as few as 2 steps (and disable guidance_scale by using a value of 1.0, as LCM is very sensitive to it and even small values lead to oversaturation):

images = pipeline(prompt_embeds=id_emb, num_inference_steps=2, guidance_scale=1.0, num_images_per_prompt=num_images).images

Note that this technique accelerates sampling in exchange for a slight drop in quality.

Start a local gradio demo

You can start a local demo for inference by running:

python gradio_demo/app.py

Arc2Face + ControlNet (pose)

We provide a ControlNet model trained on top of Arc2Face for pose control. We use EMOCA for 3D pose extraction. To run our demo, follow the steps below:

1) Download Model

Download the ControlNet checkpoint manually from HuggingFace or using python:

from huggingface_hub import hf_hub_download

hf_hub_download(repo_id="FoivosPar/Arc2Face", filename="controlnet/config.json", local_dir="./models")
hf_hub_download(repo_id="FoivosPar/Arc2Face", filename="controlnet/diffusion_pytorch_model.safetensors", local_dir="./models")

2) Pull EMOCA

git submodule update --init external/emoca

3) Installation

This is the most tricky part. You will need PyTorch3D to run EMOCA. As its installation may cause conflicts, we suggest to follow the process below:

Create a new environment and start by installing PyTorch3D with GPU support first (follow the official instructions).
Add Arc2Face + EMOCA requirements with:

pip install -r requirements_controlnet.txt

Install EMOCA code:

pip install -e external/emoca

Finally, you need to download the EMOCA/FLAME assets. Run the following and follow the instructions in the terminal:

cd external/emoca/gdl_apps/EMOCA/demos 
bash download_assets.sh
cd ../../../../..

4) Start a local gradio demo

You can start a local ControlNet demo by running:

python gradio_demo/app_controlnet.py

Test Data

The test images used for comparisons in the paper (Synth-500, AgeDB) are available here. Please use them only for evaluation purposes and make sure to cite the corresponding sources when using them.

Community Resources

Replicate Demo

Demo link by @camenduru.

ComfyUI

caleboleary/ComfyUI-Arc2Face by @caleboleary.

Pinokio

Pinokio implementation by @cocktailpeanut (runs locally on all OS - Windows, Mac, Linux).

Acknowledgements

Thanks to the creators of Stable Diffusion and the HuggingFace diffusers team for the awesome work ❤️.
Thanks to the WebFace42M creators for providing such a million-scale facial dataset ❤️.
Thanks to the HuggingFace team for their generous support through the community GPU grant for our demo ❤️.
We also acknowledge the invaluable support of the HPC resources provided by the Erlangen National High Performance Computing Center (NHR@FAU) of the Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU), which made the training of Arc2Face possible.

Citation

If you find Arc2Face useful for your research, please consider citing us:

@inproceedings{paraperas2024arc2face,
      title={Arc2Face: A Foundation Model for ID-Consistent Human Faces}, 
      author={Paraperas Papantoniou, Foivos and Lattas, Alexandros and Moschoglou, Stylianos and Deng, Jiankang and Kainz, Bernhard and Zafeiriou, Stefanos},
      booktitle={Proceedings of the European Conference on Computer Vision (ECCV)},
      year={2024}
}

foivospar/Arc2Face