A Tensorflow implementation of the paper: Reasoning structural relation for occlusion-robust facial landmark localization.
we use the Menpo project in various ways throughout the implementation.
Please look at the installation instructions at:
http://www.menpo.org/installation/
The pre-training model is coming soon.
Currently the TensorFlow implementation does not contain tracking model we did in the submitted paper, but this will be updated shortly.
# Activate the conda environment.
source activate environment-name
# Start training
python train.py --datasets='databases/lfpw/trainset/*.png:databases/afw/*.jpg:databases/helen/trainset/*.jpg'
# Track the train process and evaluate the current checkpoint against the validation set
python eval.py --dataset_path="./databases/ibug/*.jpg" --num_examples=135 --eval_dir=ckpt/eval_ibug --device='/cpu:0' --checkpoint_dir=$PWD/ckpt/train
python eval.py --dataset_path="./databases/lfpw/testset/*.png" --num_examples=300 --eval_dir=ckpt/eval_lfpw --device='/cpu:0' --checkpoint_dir=$PWD/ckpt/train
python eval.py --dataset_path="./databases/helen/testset/*.jpg" --num_examples=330 --eval_dir=ckpt/eval_helen --device='/cpu:0' --checkpoint_dir=$PWD/ckpt/train
# Run tensorboard to visualise the results
tensorboard --logdir==$PWD/ckpt
The implementation of some functions refers to the MDM project (https://github.com/trigeorgis/mdm).