Installation instructions

This project depends on the Menpo Project, which is multi-platform (Linux, Windows, OS X) and TensorFlow. As explained in Menpo's installation isntructions, it is highly recommended to use conda as your Python distribution.

Once downloading and installing conda, this project can be installed by:

Step 1: Create a new conda environment and activate it:

$ conda create -n ibugnet python=3.5
$ source activate ibugnet

Step 2: Install menpo and menpowidgets and opencv from the menpo channel:

(ibugnet)$ conda install -c menpo menpo menpowidgets opencv

Step 3: Install the TensorFlow dependencies following the installation instructions.

We have build a wheel file for the lab machines which you can install by

(ibugnet) pip install --ignore-installed --upgrade /vol/atlas/homes/gt108/pretrained_models/tensorflow-0.11.0rc0-py3-none-any.whl

Training

(TODO)

Notebook Path: /vol/atlas/homes/gt108/pretrained_models/keypoints/69_classes

Name		Name	Last commit message	Last commit date
Latest commit History 39 Commits
conv_dpm		conv_dpm
notebooks		notebooks
.gitignore		.gitignore
README.md		README.md
convert.py		convert.py
data_provider.py		data_provider.py
deblurring_train.py		deblurring_train.py
face_eval_normals.py		face_eval_normals.py
hourglass_model.py		hourglass_model.py
iasonas_model.py		iasonas_model.py
keypoints_train.py		keypoints_train.py
losses.py		losses.py
normals_train.py		normals_train.py
normals_train_photoface.py		normals_train_photoface.py
resnet_model.py		resnet_model.py
segmentation_train.py		segmentation_train.py
setup.py		setup.py
svs_model.py		svs_model.py
svs_train.py		svs_train.py
utils.py		utils.py
utils_3d.py		utils_3d.py