Classification of Men's Hairstyles on Photos from Different Angles Using Hair Segmentation

The article discusses a new classification system for haircuts, which consists of two neural networks. The user can get information about the type of hairstyle and the guide on how to do this hairstyle by photo. For the first part of the model, we use a convolutional neural network of a complex structure that highlights hair using segmentation in the photo and collects data about the parameters of the hairstyle. This allows the system to more accurately determine the type of male haircut, allowing you to use this tool in various fields of activity. The second neural network is a classifier that processes the information received from the first neural network and determines the type of haircut of the user.