TURNS HUMAN BODY INTO A TOUCH SCREEN INPUT
Presented by :Neha Pevekar
What is skinput
How it works
Principle of skinput
Skinput is named because it uses the human
Skinput, is a sensor system.
Developed by chris harrison(Mellon
Reduce gizmo accessories and multiple
What is skinput
Giving input through skin
It listen to vibrations in our
Skinput uses a series of
sensors to track where a
user taps on his arm.
Provide an always
available mobile input
Turns The body into a
The arm is an instrument
How it works
It needs bluetooth
A microchip-sized pico
An acoustic detector to
• When User tap on skin(Hand),the Bio-Acoustics &
Sensors study the sound waves.
• Variations in bone density,size and mass and the soft
tissue and joints create Acoustically different locations
• When a finger taps the skin ,several distinct forms of
acoustic energy are produced
• Londitudinal wave
• Transverse wave
When you tap your skin with your finger you generate
Tapping on soft regions of the arm create higher
amplitude transverse wave than tapping on boney
Cause internal skeletal structure to vibrate
These waves travel through the soft tissues of the arm
Joints play an important role in making tapped
locations acoustically distinct.
This makes joints behave as acoustic filters.
• Signal is sensed and worked
• This is done by wearing
• The two sensor packages
• Each contain five, specially
• piezo films, responsive to a
particular frequency range.
• Two arrays of five sensing
elements incorporated into
• Two sensor packages focus
on the arm of input
• One package was located
near the Radius other near
• Signals transmitted though
• Bio-Acoustics &
Connected to the
a system use a tiny projector to
display a screen onto your
forearm or hand
Then the menu displayed by
the Pico-Projecter on user’s
• projector display image on arm
finger tap on arm
vibrations produced and
through bones onto skin
electronic signals produced
armband in the form of music
then detected by detector in
Ten channels of acoustic data generated by three
finger taps on the forearm, followed by three taps on the wrist.
The exponential average of the channels is shown in red.
input windows are highlighted in green
To evaluate the performance of our system, 13 participants (7 people) were
These participants represented a diverse cross-section of potential ages and body
Ages ranged from 20 to 56 (mean 38.3), and computed body mass indexes (BMIs)
ranged from 20.5 (normal) to 31.9 (obese).
Three input groupings from the multitude of possible location combinations to
test were selected
These groupings, illustrated in Figure ,are of particular interest with respect to
interface design, and at the same time, push the limits of our sensing capability.
From these three groupings, five different experimental conditions are derived
,which are described as:
One set of gestures we tested had participants
tapping on the tips of each of their five fingers.
Provide clear, discrete interaction points, which
are even already well-named (e.g., ring finger).
In addition to five finger tips, there are 14
knuckles (five major, nine minor), which, taken
together, could offer 19 readily identifiable input
locations on the fingers alone.
The fingers are linearly ordered, which is
potentially useful for interfaces
acoustic information must cross as many as five
(finger and wrist) joints to reach the forearm
We decided to place the sensor arrays on the
forearm, just below the elbow.
We selected these locations for two
First, they are distinct and named parts
of the body (e.g., “wrist”).
We used these locations in three
One condition placed the sensor above the
elbow, while another placed it below.
participants repeated the lower placement
condition in an eyes-free context:
participants were told to close their
eyes and face forward, both for training
This experimental condition
used 10 locations on just the forearm
To maximize the surface area for input, we
placed the sensor above the elbow
Rather than naming the input locations, we
employed small, colored stickers to mark
we believe the forearm is ideal for
projected interface elements; the stickers
served as low-tech placeholders for
• Susceptible to
variations in body
• Prevalence of fatty
tissues and the
density/mass of bones
• Accuracy was
significantly lower for
participants with BMIs
above the 50th
To maximise the surface area for
input,we placed the sensor above
the elbow,leaving the entire
This increases input consistency
Accuracy does drop when 10 or
more locations are used
The sensor can spot many different locations
on the arm
Walking and jogging
This experiment explored the accuracy of our system.
Each participant trained and tested the system while walking and jogging on
Three input locations were used to evaluate accuracy the rate of false
positives and true positives was captured.
In both walking trials, the system never produced a falsepositive
In the jogging trials, the system had four false-positive
input events (two per participant) over 6 min of continuous
Accuracy, however, decreased to 83.3% and 60.0% for the male and female
• It is a bimanual gestures
• First had participants
tap their index,middle,
ring and pinky fingers
against their thumb ten
combined taps and
Surface and Object Recognition
• Ability to identify the
• participants to tap their
index finger against
1) a finger on their
2) a paper pad 80
3) an LCD screen.
Design and Setup
• Participant performing tasks having five
conditions in randomized order
• One sensor package rested on the biceps
• Right-handed participants had the armband
placed on the left arm
• Tightness of the armband was adjusted to be
• Experimenter walked through the input
locations to be tested
• Participants practiced the motions for one
minute with each gesture
• To convey the appropriate tap force
• To train the system, participants were
instructed to tap each location ten times
Higher accuracies can be achieved by collapsing the ten input
locations into groups.A-E and G were created using a designcentric strategy. F was created following analysis of perlocation accuracy data
• Classification accuracies for the test phases in
the five different conditions
• Rates were high, with an average accuracy
across conditions of 87.6%
• The correlation between classification
accuracy and factors such as BMI, age, and sex
Research is going on to Make armband smaller
Incorporate more devices
Extend accuracy level
No need to interact with the gadget directly.
Don’t have to worry about keypad.
Skinput could also be used without a visual interface
People with larger fingers get trouble in navigating tiny
buttons and keyboards on mobile phones. With Skinput
that problem disappears.
Individuals with visible disabilities cannot use this product.
The arm band is currently bulky.
the visibility of the projection of the buttons on the skin can
be reduced if the user has a tattoo located on their arm
If the user has more than a 30% Body Mass Index Skinput
is reduced to 80% accuracy
The easy accessibility will cause people to be more socially
• The Skinput system could display an
image of a digital keyboard on a person's
• Using Skinput, someone could send text
messages by tapping his or her arm in
• while Walking and jogging, we can listen
This system performs very well even if the
body is in motion
in the future your hand could be your
iPhone and your handset could be watchsized on your wrist.