EmotiV: Exploring Automatic Emotion Sharing through Facial Expression Recognition (FER) for Online Co-Watching

Online video watching has become prevalent, so are technologies to promote a sense of co-watching across distances. However, most co-watching technologies require active input from users (e.g. through text-based interactions) or rely on special devices. This paper presents EmotiV, a prototype designed to bring the co-watching experience to users without additional effort or devices, by automatically capturing and sharing viewers’ emotions through Facial Expression Recognition (FER). A user study with 20 participants using a comedy movie-watching scenario shows that EmotiV helped bring a sense of togetherness, aliveness and fun, and was appreciated to be more timely and authentic although with less control in comparison to traditional text-based interaction. Meanwhile, it also helped promote self-awareness and reflections, with privacy concerns to be addressed. These findings suggest that FER can serve as a lightweight and non-intrusive mechanism for augmenting remote co-watching, offering design insights for affect-aware computing to support everyday media consumption.