The code of "Rethinking Self-training for Semi-supervised Landmark Detection: A Selection-free Approach".

1. Install Python3 and PyTorch >= v1.13
2. Clone this repository.

git clone https://github.com/jhb86253817/STLD.git

3. Install the dependencies in requirements.txt.

pip install -r requirements.txt

Download the datasets from official sources, then put them under folder data/DATA_NAME/.

• 300W
• WFLW
• AFLW

1. Go to lib, run python preprocess.py data_300W to prepare 300W.
2. Configure the command in run_train.sh, then run bash run_train.sh to start training.