Skip to content

An improved version of APB2Face: Real-Time Audio-Guided Multi-Face Reenactment

Notifications You must be signed in to change notification settings

zhangzjn/APB2FaceV2

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

9 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

APB2FaceV2

Python 3.7 PyTorch 1.5.1

Official pytorch implementation of the paper: "APB2FACEV2: REAL-TIME AUDIO-GUIDED MULTI-FACE REENACTMENT".

Using the Code

Requirements

This code has been developed under Python3.7, PyTorch 1.5.1 and CUDA 10.1 on Ubuntu 16.04.

Datasets in the paper

Download AnnVI dataset from Google Drive or Baidu Cloud (Key:str3) to /media/datasets/AnnVI.

Train

python3 train.py --name AnnVI --data AnnVI --data_root DATASET_PATH --img_size 256 --mode train --trainer l2face --gan_mode lsgan --gpus 0 --batch_size 16

Results are stored in checkpoints/xxx

Test

python3 test.py 

Results are stored in checkpoints/AnnVI-Big/results

Citation

@article{zhang2021real,
  title={Real-Time Audio-Guided Multi-Face Reenactment},
  author={Zhang, Jiangning and Zeng, Xianfang and Xu, Chao and Liu, Yong and Li, Hongliang},
  journal={IEEE Signal Processing Letters},
  year={2021},
  publisher={IEEE}
}

About

An improved version of APB2Face: Real-Time Audio-Guided Multi-Face Reenactment

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published