Skip to content

kingsj0405/DIFE

Repository files navigation

Dense Interspecies Face Embedding (DIFE)

PWC PWC

An official PyTorch implementation of the paper "Dense Interspecies Face Embedding".

Dense Interspecies Face Embedding
Sejong Yang1, Subin Jeon1, Seonghyeon Nam2, Seon Joo Kim1
1Yonsei University, 2York University
in NeurIPS 2022

image

Features

  • Inference for single image
  • Training
  • Experiment
    • Interspecies keypoint transfer
    • Interspecies face parsing
  • Demo with streamlit

Setup

Docker

  • Build image
$ cd docker
$ docker build -t dife .
$ cd ..
  • Run docker container and access
$ docker run -ti -d --gpus=all --name=dife -v .:/workspace --ipc=host dife
$ docker exec -ti dife /bin/bash
# (Do something)

Pre-trained model

# python download.py --key model-human+dog

Test data

# python download.py --

Usage

Inference for single image

# python inference.py --image_path demo/human_000001.png
# python inference.py --image_path demo/dog_000001.png

image

Citation

@inproceedings{yang2022dife,
  title={Dense Interspecies Face Embedding},
  author={Sejong Yang and Subin Jeon and Seonghyeon Nam and Seon Joo Kim},
  year={2022},
  booktitle={NeurIPS},
}

Reference

About

Dense Interspecies Face Embedding (NeurIPS 2022)

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published