Data for: Face analysis through semantic face segmentation

Published: 31 January 2019| Version 2 | DOI: 10.17632/8yjv7b3hgr.2
Sergio Benini, Khalil Khan, Massimo Mauro, Riccardo Leonardi, Pierangelo Migliorati


The FASSEG repository is composed by three datasets containing face images useful for training and testing automatic methods for the task of face segmentation. Two datasets, namely frontal01 and frontal02, are specifically built for performing frontal face segmentation. Frontal01 contains 70 original RGB images and the corresponding roughly labelled ground-truth masks. Frontal02 contains the same image data, with high-precision labelled ground-truth masks. The third dataset, namely Multipose01, contains more than 200 faces in multiple poses and the corresponding ground-truth masks. For all face images, ground-truth mask data are labelled on six classes (mouth, nose, eyes, hair, skin, and background). All images of each dataset come organized in two subfolders - train and test – for enabling comparisons between different segmentation methods on the same datasets.



Image Processing, Content-Based Image Analysis