Loading…
Capacitive Measurement of Facial Activity Intensity
The intensity measurement of facial muscle activity can be used in several applications such as human-computer interaction and behavioral science. A new method for the intensity measurement is presented. It is based on a contactless, capacitive measurement of the movements that the facial activity p...
Saved in:
Published in: | IEEE sensors journal 2013-11, Vol.13 (11), p.4329-4338 |
---|---|
Main Authors: | , , , , , , |
Format: | Article |
Language: | English |
Subjects: | |
Citations: | Items that this one cites Items that cite this one |
Online Access: | Get full text |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Summary: | The intensity measurement of facial muscle activity can be used in several applications such as human-computer interaction and behavioral science. A new method for the intensity measurement is presented. It is based on a contactless, capacitive measurement of the movements that the facial activity produces. The muscles responsible for raising the eyebrows, lowering the eyebrows, raising the mouth corners, and pulling down the mouth corners are measured simultaneously with the capacitive method and electromyography (EMG) during controlled experiments. Each muscle is activated by 10 participants at three different intensity levels (low, medium, and high), 10 repetitions at each level. The capacitive intensity values are in good agreement with the ones registered with the EMG: average mean absolute errors are between 7% and 12% of the observed intensity range. However, compared with the EMG, the capacitive intensity values are noticed to have offsets that may be partly caused by the measurement itself and partly by the EMG reference. As a result, the measurement may require a calibration for more intensity values than just the maximum. In the capacitive method, it is also required to distinguish between the muscle activations originating from the same facial regions to determine which activation is taking place. This is done with an almost perfect performance by using hierarchical clustering to cluster the intensity values. |
---|---|
ISSN: | 1530-437X 1558-1748 |
DOI: | 10.1109/JSEN.2013.2269864 |