Method, server and program for cognitive and speech impairment detection based on temporal visual facial features

摘要

PROBLEM TO BE SOLVED: To provide a system, a method, and a program for generating predictive information that predicts a user's state based on an input video by using a visual facial feature of the user. A computer-implemented method of assessing a user's condition is provided that generates a facial landmark on a received input video to define and define points associated with a region of interest of the user's face. Define the duration of the conversation based on the location of each of the defined points, measure pause frequency, repetitive patterns, and vocabulary diversity during the conversation and aggregate the measurements to predict the user's state Generating prediction information. [Selection diagram] Fig. 5