Zoom meeting video

First we chop the video into frames (due to computational burden) only collected a handful of frames. Then the frames are posted on windows azure server and analyzed using facial recognition software that has already been trained. Here are the results for some of the frames.

Frame 1

Number of faces: 2.
gen age anger contempt disgust fear happiness neutral sadness surprise
male 37 0 0.002 0 0 0.029 0.968 0 0
female 29 0 0.001 0 0 0 0.8 0.199 0

Frame 2

Number of faces: 2.
gen age anger contempt disgust fear happiness neutral sadness surprise
female 28 0 0 0 0 0 0.995 0.003 0.001
male 35 0 0 0 0 0.001 0.998 0 0

Frame 3

Number of faces: 0.

Frame 4

Number of faces: 0.

Frame 5

Number of faces: 1.
gen age anger contempt disgust fear happiness neutral sadness surprise
female 9 0 0.002 0 0 0.017 0.975 0.002 0.004

Frame 6

Number of faces: 0.

Frame 7

Number of faces: 0.

Frame 8

Number of faces: 0.

Frame 9

Number of faces: 0.

Frame 10

Number of faces: 0.

Frame 30

Number of faces: 2.
gen age anger contempt disgust fear happiness neutral sadness surprise
male 36 0 0.014 0 0 0.016 0.97 0.001 0
female 31 0 0 0 0 0 0.992 0.008 0

Frame 38

Number of faces: 2.
gen age anger contempt disgust fear happiness neutral sadness surprise
female 29 0 0 0 0 0 0.896 0.103 0
male 39 0 0.003 0 0 0.302 0.694 0 0

Frame 39

Number of faces: 2.
gen age anger contempt disgust fear happiness neutral sadness surprise
male 38 0 0 0 0 1 0 0 0
female 30 0 0.007 0 0 0.886 0.107 0 0

Frame 40

Number of faces: 2.
gen age anger contempt disgust fear happiness neutral sadness surprise
female 30 0 0 0 0 0 0.998 0.002 0
male 35 0 0.008 0 0 0.022 0.968 0.002 0