The Frontiers of HCI Jim Warren derived from lots of places and with thanks to Beryl Plimmer Haptics, Smell and Brain Interaction.

The Frontiers of HCI

Jim Warrenderived from lots of places and with thanks to Beryl Plimmer

Haptics, Smell and Brain Interaction

Learning Outcomes

• Describe haptics in terms of– Human perception– Applications– Devices

• Describe application of eye tracking and visual gesture recognition

• Describe the exploration of – Olfactory detection and production– Brain wave detection

The Human Perceptual System

• Physical Aspects of Perception– Touch (tactile/cutaneous)• Located in the skin, enables us to feel – Texture – Heat– Pain

– Movement (kinesthetic/proprioceptive)• The location of your body and its appendages• The direction and speed of your movements

1-3

Physical Aspects of Perception• Proprioception– We use sensation from our joints (e.g. their angles) and

our muscles (e.g. strain) to determine the position of our limbs and perceive body position

• Combine with vestibular system (inner ear, balance) to perceive motion, orientation and acceleration– This combination is sometimes called the

kinaesthetic sense.

1-4

Mobile devices

• Phone output– Vibrate – silent alert. • These can be used like earcons – different signals for

different events• Does your phone have different alerts?

– Can you tell the difference?

1-5

Mobile devices

• Phone input– Touch screens• See previous lecture

– Accelerometer - shaking actions • Inconsistent interactions, high error rates

– Passive input• GPS• Altimeter, Temperature, Humidity• Specialised fitness or medical monitors

1-6

Fitbit Flex with sleep

tracker

Using Haptics in Interaction Design

• ImmersiveTouch™ high fidelity surgical simulators

1-7

3D view with hi-res graphics

Realistic surgical instrument attached

to force feedback controller

Using Haptics in Interaction Design• Medical Uses– Surgeon controls ‘robot’ with zoomed view and

automated enhancements over manual surgery (e.g. greater range of motion than human wrist, tremor reduction)

• Most famous is da Vinci Surgical System– Over 200,000 operations in 2012– Mostly prostate, uterine

and heart valve (i.e.delicate stuff)

1-8

Using Haptics in Interaction Design

• The GuideCane (Ulrich and Borenstein, 2001)

1-9

Force Feedback Displays

– Manipulator Gloves

1-10

CyberForc

e

CyberGraspCyberGlove

II

Motion capture

Force feedback – can ‘hang your hand’ on a virtual steeringwheel

Feel size and shape of

virtual object

Desktop Haptic Devices– SensAble PHANTOM (now by GeoMagic)• Closest to a commodity force feedback tool

1-11

PHANTOM Premium

6-degrees (3

translational,

3 rotational)

McSig – Beryl’s work with visually

impairedhttp://dl.acm.org.ezproxy.auckland.ac.nz/citation.cfm?id=1993060.1993067&c

oll=DL&dl=ACM&CFID=333467105&CFTOKEN=52399920

Eye tracking

1-12GP3 eye tracker by Gazepoint

• Most consumer-friendly modern method uses infrared (IR) light reflected off of different parts of the eye to detect angle of gaze– 1st-4th Purkinje images are from outer and inner surfaces of

cornea (1, 2) and outer and inner surfaces of lens (3, 4)– Measures of these angles from multiple locations,

combined with measure of head position,allow estimation of gaze point on screen*

* Chi Jian-nan ; Zhang Peng-

yi ; Zheng Si-yi ; Zhang Chuang

; Huang Ying, Key Techniques of

Eye Gaze Tracking Based on

Pupil Corneal Reflection, IEEE

Intelligent Systems, 2009

Eye tracking applications• Using eye tracking to estimate gaze point over time provides rich

insight into how users consume a visual presentation– E.g. what do they look at on a Web page and for how long?

• Rather more limited as an input method – Careful control of eye gaze

to act as pointer can resultin eye strain*

– But still useful for peoplewith disabilities

– Can use dwell time toindicate click, but error-prone

• Room for further research incombining with other inputmethods

1-13*http://www-edc.eng.cam.ac.uk/~pb400/

Papers/4_pbiswas_JAT11a.pdf

Visual gesture recognition• OpenKinect – sort of pirate community making API to use

Xbox Kinect hardware widely• Makes skeleton from video and learns gestures for control

– OK with broad gestures and good contrast to background– Kinect reasonably well received as a video game enhancer– Other apps being explored (e.g.

surgeon to work computerwith sterile hands)

1-14http://support.xbox.com/en-NZ/xbox-one/kinect/common-gestures

Olfactory - Odour/ Smell

• Smell is essentially our ability to detect specific chemical particles in the air

• We can detect about 4000 different smells• And they can be combined in millions of

different ways • Smell is very deep

in our animal brain

1-15

Smell – current research • Using sounds and smell

signatures to aid recallof and affinity withindividuals

• A wire in the glassesheats 8 perfumes torelease a scent

• Yongsoon Choi, Rahul Parsani, Xavier Roman, Anshul Vikram Pandey, and Adrian David Cheok. 2012. Sound perfume: building positive impression during face-to-face communication. In SIGGRAPH Asia 2012 Emerging Technologies (SA '12). ACM, New York, NY, USA, , Article 22 , 3 pages. DOI=10.1145/2407707.2407729 http://doi.acm.org.ezproxy.auckland.ac.nz/10.1145/2407707.2407729

• Read http://www.newscientist.com/article/mg21228415.800-glasses-emit-personal-sound-and-smell.html#.Uz4nzI3HlSF

• See video: https://www.youtube.com/watch?v=4-UTFNaesTc

1-16

Technology of Odour• Input

– Detecting particular chemicals is possible • Drug/ explosive sniffers

– Detecting the range of smells in anything like human terms is extremely difficult task

• Output– Manufacturing particular smells possible (e.g. ‘freshly baked

cookies’– Active generation of a range of smells very difficult, but

choosing a single smell to assert branding and positive association for a retail outlet or such is already done (see http://www.scentair.com/why-scentair-news-press/smells-sell-as-hard-nosed-traders-discover/)– Actually not that different than the conventional use of perfume to

create an almost-subliminal association for one’s partner– Also similar to branding with corporate colours

1-17

Brain Computer Interaction

• Detecting the brain waves and interpreting• From outside the skull – not very accurate• Inside the skull – accurate but invasive http://www.youtube.com/watch?v=ogBX18maUiM

1-18

Reading nerve signals from brain to muscles

• Application: Motor Disabilities– HAL-5 (Hybrid Assistive Limb), CYBERDYNE Inc.

www.cyberdyne.jp

1-19

1-20

EEG and Visually Evoked Potentials (VEP)

• Electroencephalography (EEG) is the recording of electrical activity along the scalp

• Patterns (in shape, or as strobing) coming into the eye can translate to measurable signals on the EEG (VEP)– However, there are many sources of noise, including

blinking– And it’s not a rapid-response thing (usually analyse period

200-500ms following onset of visual stimulus) – See http://webvision.med.utah.edu/book/electrophysiology/visually-evoked-potentials/

1-21

Summary • Describe haptics in terms of

– Human perception • Touch, proprioception, kinaesthetics

– Applications• Surgery (training or actual), assistive technology

• Describe applications of eye tracking and visual gesture recognition– Eye tracking: user studies, assistive; Visual gesture: games

• Describe the exploration of – Olfactory detection and production

• Detection of specific chemicals possible • Production of a limited range of scents

– Brain wave detection• Awkward set up and use through EEG, but a boon to those who need it • Fairly limited interaction without surgery