ffmpeg
cuda (v8.0)
For face alignment:
scikit-image
dlib (v18.18)
numpy
scipy
cd matconvnet_gen
addpath matlab
vl_compilenn('enableGpu', true, ...
'cudaRoot', '/usr/local/cuda-8.0', ...
'cudaMethod', 'nvcc') ;
vl_testnn('gpu', true)
Download model:
sh download_model.sh
Run demo in MATLAB:
run_demo
cd face_detection
y = baseface('../data/obama.jpg');
5 aligned identity images are then concatenated channel-wise. See faceimg.mat for examples.
Please cite the paper below if you make use of the demo.
@InProceedings{Chung17b,
author = "Chung, J.~S. and Jamaludin, A. and Zisserman, A.",
title = "You said that?",
booktitle = "British Machine Vision Conference",
year = "2017",
}