Making facial expressions of emotions accessible for visually impaired persons

H Buimer, T Van der Geest, A Nemri… - Proceedings of the 19th …, 2017 - dl.acm.org
Proceedings of the 19th International ACM SIGACCESS Conference on Computers …, 2017dl.acm.org
One of the big problems visually impaired persons experience in their daily lives, is the
inability to see non-verbal cues of conversation partners. In this study, a wearable assistive
technology is presented and evaluated which supports visually impaired persons with the
recognition of facial expressions of emotions. The wearable assistive technology consists of
a camera clipped on spectacles, emotion recognition software, and a vibrotactile belt with six
tactors. An earlier controlled experimental study showed that users of the system improved …
One of the big problems visually impaired persons experience in their daily lives, is the inability to see non-verbal cues of conversation partners. In this study, a wearable assistive technology is presented and evaluated which supports visually impaired persons with the recognition of facial expressions of emotions. The wearable assistive technology consists of a camera clipped on spectacles, emotion recognition software, and a vibrotactile belt with six tactors. An earlier controlled experimental study showed that users of the system improved significantly in their ability to recognize emotions from validated stimuli. In this paper, the next iteration in testing the system is presented, in which a more realistic usage situation was simulated. Eight visually impaired persons were invited to participate in conversations with an actor, who was instructed not to exaggerate his facial expressions. Participants engaged in two 15-minute mock job interview conversations, during one of which they were wearing the system. In the other conversation, no assistive technologies were used. The preliminary results showed that the concept of such wearable assistive technologies remains feasible. Participants within the study found it easy to learn and interpret the vibrotactile cues, which was also shown in their training performance. Furthermore, most participants could use the vibrotactile cues, while being able to stay engaged in the conversation. Nevertheless, some improvements are needed before the system can be used as assistive technology. The accuracy of the system was negatively affected by the lighting and movement conditions present in realistic conversations, compared to the controlled experiment condition. Furthermore, participants requested developments to improve the wearability of the system.
ACM Digital Library