SDM

Supervised Descent Method for Face Alignment using Python

• First, download the dataset used by this project, which is the lfpw dataset. Link: https://pan.baidu.com/s/1jIJNg2q pwd: f36i

• Second, get the data from data.tar that just downloaded

• Third, put main.py in the same directory with the data folder, then run the main.py

• For the first time, the main.py will run the train with a parameters, and after training process, you will get a train_data.mat file in the current directory. If you run the main.py with a train_data.mat file already there, the main.py will load the R,B,I from the file without the training process.

• After you have get the R,B,I, you simply run the function test_after_run_main(n) to test the number nth image in the testset.