Fig. 2 | Scientific Reports

Fig. 2

From: Recognition of anxiety and depression using gait data recorded by the kinect sensor: a machine learning approach with data augmentation

Fig. 2

Real-time display of the Kinect sensor during walking samples for subject recognition. Parts (a),(k), and (l) show instances of joint distortion: part (a) was removed due to the loss of position in the ankle, knee, and thigh joints of the right leg, while parts (k) and (l) were excluded due to loss of position in the upper and lower trunk joints during preprocessing. The remaining parts—(b), (c), (d), (f), (g), (h), (i), and (j)—represent the frames selected for feature extraction.

Back to article page