Overview / Usage

The objective of this task is to predict keypoint positions on face images. This can be used as a building block in several applications, such as:

• tracking faces in images and video
• analysing facial expressions
• detecting dysmorphic facial signs for medical diagnosis
• biometrics / face recognition

Methodology / Approach

I primarily used deep learning in this project. The data was limited so I used data augmentation but traditional data augmentation techniques won't work here because they might not change the landmarks of the augmented images accordingly. To resolve this issue I used `imgaug`. Forr deep learning-related tasks, I used the high-level library of TensorFlow - Keras.