Dr. Mónica Gracia Villar, a researcher graduated from the Universidad Internacional Iberoamericana (International Iberoamerican University) (UNIB), together with other researchers, among them, Dr. Carmen Lili Rodríguez of the Universidad Europea del Atlántico (European University of the Atlantic) (UNEATLANTICO), propose a new methodology for detecting emotions through facial expressions involving occlusions and inclination.
In recent years, emotion detection has been gaining relevance due to the fact that robots are expected to have the ability to speak with human-like emotions in the near future. In addition, efficient emotion analysis is gaining popularity in surveillance imaging, home automation, computer games, stress treatment, observation of depressed patients, and psychoanalysis studies.
Although various methods have been proposed, obstacles such as occlusion, variations in illumination, and changes in physical appearance have been encountered. This new methodology proposes a new architecture design of a convolutional neural network (CNN) for the FER system and has five convolution layers; improvements were made in the function map to achieve higher accuracy and determine more emotions. In this way, the obstacles of occlusion and illumination are solved.
If you want to learn more about this fascinating study, click here.
To read more studies, visit the UNIB repository.
Finally, the Universidad Internacional Iberoamericana (International Iberoamerican University) (UNIB) offers a Master's Degree in Project Design, Administration and Management, with a specialty in Innovation and Product.
This program aims to train professionals with high social and ethical commitment, providing them with the knowledge to develop innovative projects, such as this new methodology proposed by the researchers. Therefore, UNIB congratulates the researchers for their noble work and innovative spirit.