Analysis of databases of facial expressions for the automatic identification of learning-centered emotions

Abstract

This work presents the analysis of the state of the art of facial expressions databases for the automatic identification of learning-centered emotions. Obtaining data for automatic recognition processes in a specific context is essential for their success. Thus, this project begins by reviewing the information available to carry out the training and classification stages of emotions with the proposed computational techniques. The search activities of the databases of facial expressions that capture learning-centered emotions are described. These activities were part of the stages of the work methodology to recognize students' emotions while they carried out online learning activities. This allowed justifying the creation of the database, formalizing a protocol from its capture to its digitization.