Continuous Engagement Estimation based on Gaze Estimation and Facial Expression Recognition

초록

Existing engagement estimation methods only show discrete estimation results according to the degree of engagement, and do not express fine-grained state information of the object. In this paper, we propose a continuous engagement estimation framework using estimated gaze direction and facial expression information. Also, we build a video-based engagement estimation database and succeeded in qualitatively verifying the performance of our framework.

키워드