Gustavo Correa

MIT Department: Media Arts and Sciences

Undergraduate Institution: University of California, Riverside

Faculty Mentor: Joseph Paradiso

Research Supervisors: Nan Zhao, Ariel Ekblaw

Websites: Personal Website, LinkedIn, Github

2018 Research Poster


My name is Gustavo Jaime Correa and I come from Fontana, California. I am a fourth year electrical engineering major at the University of California, Riverside with an interest in developing robots for Earth and space. Outside of school, I am either conducting robotics research under Professor Konstantinos Karydis on multi-agent systems for bio-inspired small legged robots, or exploring my artistic side through personal projects and music festivals. During my time at MIT, I’ve been able to reinforce my interests on advancing to a higher degree with the goal of obtaining a Ph.D. and become a leading roboticist in my field. Through robotics research,  I hope to extend the capabilities of an individual’s body and mind so that we can reach and explore the final frontier.

2018 Research Abstract

Using Facial Thermal Variations for Increased Personalization of a Mobile Mediated Atmosphere

Gustavo Correa1, Nan Zhao2 and Joe Paradiso3

1Department of Electrical and Computer Engineering, University of California, Riverside

2, 3Department of Media Arts and Sciences, Massachusetts Institute of Technology

Mobile mediated atmospheres are systems that digitally control characteristics of the ambient environment using sound, lighting, and video projection to affect cognitive performance, behavior, and physiology. These systems are able to change mood, increase productivity, and improve the overall well being of an individual in a work environment. Presently, non mobile versions of these systems use mostly wearable methods to measure facial expressions and physiological changes. These measurements are then processed to develop a model of the individual’s mental, emotional, and physiological state. To implement mediated atmospheres into mobile robots, new contact-free mechanisms are needed to measure cognitive load and physiology to create a robust model used in real-world applications. Our solution for contact-free cognitive load estimation is to use thermal cameras to measure facial thermal variations (FTV) on the forehead and nose regions. I conducted within-subject experiments to observe FTV in individuals performing two cognitive tasks under different noise factors including ambient temperature, lighting, sound, and varying video projection sharpness. Cognitive load estimations from FTV measurements provide additional input to mobile mediated atmosphere systems designed to personalize and regulate environmental conditions in the ambient atmosphere. Leveraging thermal cameras for contact-free cognitive estimation will enable increased personalization of mediated atmospheres, assistive robots, and attention-aware interfaces.