Seat Comfort Evaluation Using Face Recognition Technology

Abstract: One of the difficulties inherent to comfort assessment is to translate comfort perception into quantifiable variables in order to measure it and use this result to improve seat comfort. This study describes the opportunities of using facial expressions recognition technology to compare comfort perception of two aircraft seats installed in a representative environment. Facial expressions are one of the most apparent ways to capture emotions and it is known that there are six basic emotions which are universal throughout human cultures: fear, disgust, anger, surprise, happiness and sadness. Twenty-one subjects (18 males and 3 females) participated in this experiment and have their faces recorded while using the seats and being asked some questions. The recordings obtained were posteriorly analyzed by Emotion Research Lab facial recognition technology to obtain an emotional analysis of the facial expressions displayed by the participants during the experiment. The facial expressions recognition software of Emotion Research Lab captures the facial micro expressions and uses them to predict the behavior of the participants through the calculation of different metrics such as activation, engagement, satisfaction, valence, relevance and enjoyment. The results showed that seat 1 was better rated by participants and had emotional congruence with their answers. The most important finding was that even subtle differences in seats could be perceived in participants’ emotions, suggesting that the use of facial expressions recognition technology to compare comfort perception of aircraft seats is viable and should be better explored during seat development process.