Virtual Facial Expression Dataset UIBVFED

The dataset is composed of 640 facial images from 20 virtual characters each creating 32 facial expressions. The avatars represent 10 men and 10 women, aged between 20 and 80, from different ethnicities. Expressions are classified by the six universal expressions according to Gary Faigin classification.

MASEPESmile Data Base

MASEPESmileDB example

The database consists of 91 videos in an environment with controlled lighting and with the same background colour on all shots. Seven subjects with acting experience performed in the videos. Each subject was asked to perform 13 expressions, starting at the neutral expression. We have shown the actors the image of the expression that appears in Faigin’s work and have contextualized a situation where that facial expression would occur. Each actor reproduces each type of laughter or smiles only once according to the explanation and visual information received. The recording is done once the actor has perfectly understood the type of expression required. In case of doubt, we recorded several shots. Recording sessions are performed with only one actor at a time, so they are not influenced by the performances of others