Fig. 5From: Predicting students’ attention in the classroom from Kinect facial and body featuresExamples of final feature set for test person 6. Image (a) represents a single frame of Kinect video with visualization of Kinect data including face rectangles, face points, gaze vectors, and body skeletons. Graph (b) shows reference attention level, and graph (c) shows seven normalized and smoothed feature signalsBack to article page