This is the Repository of our paper of Robust Facial Alignment with Internal Denoising Auto-Encoder.
Requirements :
- MTCNN requirements : https://github.com/DuinoDu/mtcnn
- Tensorflow GPU : https://www.tensorflow.org/install/install_linux
- Pytorch GPU : https://pytorch.org/
- Other packages, can be installed by cloning my environment file on src : env.yml
This Repository hold the Facial Localiser from this repo : https://github.com/deckyal/RT
Example of the facial alignment
usage :
python DenoiseLocalise.py
Change : config.py :
- img_name = 'indoor_120.png' #Your image name
- trySynthetic = True/False #Whether to provide the demo on the synthetic or real data(300VW)
- image_directory = 'ex_images' #the directory of the image
- cl_type = 0#0,1,2 the type of denoising.
Citation :
D. Aspandi, O. Martinez, F. Sukno, X. Binefa, Robust facial align-ment with internal denoising auto-encoder, in: 2019 16th Confer-ence on Computer and Robot Vision (CRV), 2019, pp. 143โ150.doi:10.1109/CRV.2019.00027.