Electronics keep getting smaller, but human fingers don’t. This leads to a real challenge with highly-embedded wearable computers. Sure, voice command has come a long way, but it has its own challenges. You might not want to verbally command your Borg implants in some situations. Maybe you need to be quiet. Or perhaps you are worried about accidentally triggering the device. Researchers in Germany want to monitor facial expressions instead. So to snap a picture, you might wink and to fast forward your movie playing on the inside of your eyelids, perhaps you’d look to the right twice. You can see a video presentation about the paper, below.
The paper looks at several different methods to read facial movements. Some were pretty intrusive. However, a promising technique used Ear Field Sensing (EarFS). An earplug with an electrode senses electrical changes in the ear canal resulting from facial muscle movement. Other techniques examined included electromyography, capacitive sensing, and a different form of electrical field sensing.
The team identified 25 different facial expressions, although they settled on the best five for each technology. It seems that people start forgetting what to do as the command list passes beyond about seven gestures.
The circuit to sense the ear was created in Eagle, and appears in the paper. There are some very high resistances involved along with an instrumentation amplifier and two dual op amps. The circuit has a few options for working single-ended or differential with different filters. Amusingly, for testing the researchers stuck a pinky in their ears to feel what the various facial expressions did inside the ear canal.
Testing with the devices included in a static setting and while in motion. None of the technologies were perfect, but the data does show that it is possibly useful if you could miniaturize and package the circuits in a more consumer-friendly fashion.
We thought about using OpenCV to look at facial expressions, but while that’s one thing for a robot, it would be hard to do with a device you were wearing. We wonder if this technique would be less useful for command and more useful for figuring out how ticked off you were.
28bits of C resolution? I’m curious how many are actually usable.
…also I predict increase in socially awkward interaction, similar to bluetooth headset, when someone talking in your direction, but not to you. Now add to that smiles and head nods :)
Totally off topic, but are those switchable jumpers or just single dipswitches? The pcb shows 2 and 3 pins per switch.
I haven’t seen those before, (thanks for mentioning that), but to me they look like Single pole Double Throw miniature
switches.
found them
https://www.digikey.com/product-detail/en/apem-inc/NK236/679-1860-ND/1795659
Thanks!
I see no big need to capture facial expressions for anybody but 3rd party ‘actors’ with nasty intentions. Ads or worse.
“A psychiatrist is a person, who, when a pretty woman enters a room, watches everybody else.”