BOVIFOCR Instructions

1. Main requirements

• Python==3.9
• CUDA==11.2
• numpy==1.23.1
• mxnet==1.9.1
• torch==2.3.0
• torchvision==0.18.0
• pytorch-lightning==1.7.1
• opencv-python==4.9.0.80

2. Create environment

CONDA_ENV=dcface_synthetic_face
conda create -y --name $CONDA_ENV python=3.9
conda activate $CONDA_ENV

conda env config vars set CUDA_HOME="/usr/local/cuda-11.2"; conda deactivate; conda activate $CONDA_ENV
conda env config vars set LD_LIBRARY_PATH="$CUDA_HOME/lib64"; conda deactivate; conda activate $CONDA_ENV
conda env config vars set PATH="$CUDA_HOME:$CUDA_HOME/bin:$LD_LIBRARY_PATH:$PATH"; conda deactivate; conda activate $CONDA_ENV

3. Clone this repository and install requirements

cd ~
git clone https://github.com/BOVIFOCR/dcface_synthetic_face.git
cd dcface_synthetic_face
./install.sh

• If it fails, try to run pip3 install -r requirements.txt and download model weights from the link

Using Pre-Trained Models

- Creating Synthetic ID Images (New Subjects)

cd dcface/stage1/unconditional_generation
python unconditional_sampling.py \
      --attention_resolutions 16 \
      --class_cond False \
      --diffusion_steps 1000 \
      --num_samples 16 \
      --batch_size 8 \
      --image_size 256 \
      --learn_sigma True \
      --noise_schedule linear \
      --num_channels 128 \
      --num_head_channels 64 \
      --num_res_blocks 1 \
      --resblock_updown True \
      --use_fp16 False \
      --use_scale_shift_norm True \
      --timestep_respacing 100 \
      --down_N 32 \
      --range_t 20 \
      --save_dir unconditional_samples

• 16 new images will be saved into the folder dcface/stage1/unconditional_generation/unconditional_samples

- Stylize images

cd dcface/src
python synthesis.py --id_images_root sample_images/id_images/sample_57.png --style_images_root sample_images/style_images/woman

• The image sample_57.png will be stylized from images sample_5.png, sample_14.png, sample_20.png, sample_45.png, sample_63.png, sample_69.png and saved into the folder dcface/generated_images/dcface_5x5/id:sample_57/sty:list_woman

Train Mixer Model

Dataset preparation

• Download casia webface dataset from insightface
  • Place it under $DATA_ROOT (ex: /data/).
  • ex: /data/faces_webface_112x112

Download model weights

• Download all pretrained weights from the link
• Place the pretrained_models directory under dcface (same level as src)

Run

cd dcface
bash src/scripts/train.sh

DCFace: Synthetic Face Generation with Dual Condition Diffusion Model

Official repository for the paper DCFace: Synthetic Face Generation with Dual Condition Diffusion Model (CVPR 2023).

• Arxiv: https://arxiv.org/abs/2304.07060
• Main paper: main.pdf
• Supplementary: supp.pdf

Pipepline

ID Consistent Generation

Installation

1. one-liner : install.sh

If above fails, then

pip install -r requirements.txt
# and download model weights from the link below

Image generation

Sample Code with given example images

We provide the sample code to generate images with the pretrained weights. The sample aligned images are provided in the repository.

• Download the pretrained weights from the link
• Place the pretrained_models directory under dcface (same level as src)
• Run

cd dcface/src
python synthesis.py --id_images_root sample_images/id_images/sample_57.png --style_images_root sample_images/style_images/woman

Optional

One can also generate new subject images and prepare custom style images.

1. Creating ID Images (New Subjects)

Unconditional ID image generation is done in dcface/stage1/unconditional_generation Take a look at the README.md in that directory for instructions on how to generate new ID images.

2. Creating Style Images

Any aligned images can serve as style images. We provive some sample images in sample_images/style_images directory. For anyone who wants to use their own style images, one should align the images first. Take a look at the README.md in dcface/stage1/style_bank directory for instructions on how to align images.

3. Using 1. and 2. to generate dataset

Assuming that you followed 1. and 2. you will have an id_image and style_images directory. For the sake of explaination, let's say

• ID image is <Project_root>/dcface/stage1/unconditional_generation/unconditional_samples_aligned/00011.png
• Style directory is <Project_root>/dcface/stage1/style_bank/style_images/raw_aligned Then to combine these run by pointing at these paths,

cd dcface/src
python synthesis.py \
        --id_images_root <Project_root>/dcface/stage1/unconditional_generation/unconditional_samples_aligned/00011.png \
        --style_images_root <Project_root>/dcface/stage1/style_bank/style_images/raw_aligned

The result will be saved at <Project_root>/dcface/generated_images/

Training generator

Dataset preparation

• Download casia webface dataset from insightface
  • Place it under $DATA_ROOT (ex: /data/).
  • ex: /data/faces_webface_112x112

Download model weights

• Download all pretrained weights from the link
• Place the pretrained_models directory under dcface (same level as src)

Run

cd dcface
bash src/scripts/train.sh

Dataset Release

DCFace synthetic dataset can be downloaded from link

• dcface_0.5m_oversample_xid.zip
• dcface_1.2m_oversample_xid.zip

The format of the downloaded file is in rec format.

• you can convert it to png using the script.
• rec file will be useful for the face recognition training script provided in the repository. (to be released soon)

cd dcface/convert
python record.py --rec_path <path_to_rec_file> --save_path <path_to_save_png>
# ex
# <path_to_rec_file> : dcface_0.5m_oversample_xid/record
# <path_to_save_png> : dcface_0.5m_oversample_xid/images

Training Face Recognition Model

• to be released soon