This toolchain implements a simple face anti-spoofing algorithm which works as a binary classifier. The algorithm expects video files as input and processes the data from a training, development and test set and trains a binary classifier given the training data. The real accesses are considered as the positive, while the attacks as the negative class.

The toolchain consists of the following steps:

1. Preprocessing of the input files. This step depends on two inputs from the database: the input video file and face location file. This step could be, for example, cropping of the faces from each frame of the input video and normalizing them by size.
2. Feature extraction. This step extracts relevant features from the input frames.
3. Training. This step trains a binary classifier based on the training video.
4. Classification. This step classifies the videos in the development and test set.
5. Evaluation. This step calculates error rates and plots different performance curves.

The toolchain is based upon the work presented in [Chingovska12]. All the algorithms presented there can be readily implemented and used with this toolchain.