Technology Sharing

한어Русский языкEnglishFrançaisIndonesianSanskrit日本語DeutschPortuguêsΕλληνικάespañolItalianoSuomalainenLatina

1. Download the github project 3dmm_cnn-master
    https://github.com/anhttran/3dmm_cnn.git
A project for 3D face modeling from a single image using deep neural networks. End-to-end code that estimates 3D shape and texture directly from image intensity; uses a regressed 3D face model to estimate head pose and expression from detected facial landmarks.
    
2. Download relevant data
1. Download the 3DMM_CNN model and move the 3 files to the folder CNN
    http://www.openu.ac.il/home/hassner/projects/CNN3DMM
        3dmm_cnn_resnet_101.caffemodel
        deploy_network.prototxt
        mean.binaryprotoCNN
2. Download the Basel face model and move it into the folder 3DMM_model
    https://faces.dmi.unibas.ch/bfm/index.php?nav=1-2&id=downloads
    01_MorphableModel.mat
3. Get the 3DDFA expression model to the folder 3DMM_model
http://www.cbsr.ia.ac.cn/users/xiangyuzhu/projects/3DDFA/Code/3DDFA.zip may be invalid
    https://onedrive.live.com/?authkey=%21AGNgYRrq6OQPZE4&id=1B5D487B5A28CEFD%213218&cid=1B5D487B5A28CEFD 可用