Interaktion i Virtual Reality Jesper Kjeldskov CVMT

1
Interaktion i Virtual Reality Jesper
KjeldskovCVMT
2
Interaktionsprojektet

• Samarbejde mellem Staging og VRCN
• 6 måneders fokus på interaktion i VR med
  udgangspunkt i VRCNs faciliteter
• Skabelse af fundament for arbejde med interaktion
  i VR ved VRCN
• Fokus på både teknologi og anvendelse

3
Dagens fokus

• Introduktion til interaktion i VR
• Problemstillinger ift. interaktion i VR
• Alternative løsninger
• Fordele
• Ulemper
• Mulighederne for at lave interaktive VR
  applikationer ved VRCN i dag

4
Myths of interaction i VR

• There exist one optimal interaction technique
  for virtual reality
• Different displaytypes should always use the
  same interaction technique

5
Reality
A good interaction technique depends on the
relations between

• Inputdevices
• Display type
• Application context
• Type of application
• Users skills
• Etc.

6
Desktopdevices
Spaceball/Spacemouse
7
Positiontrackers
Electromagnetic (Polhemus, Assension osv.)
8
Positiontrackers
Computervision
9
Combined devices
Wanda, TrackMan
10
Gloves
CyberGlove, 5dt glove
11
Interactiondevices of tomorrow?
brainfingers
12
Display types
13
Categorizing display types

• Full immersive displays
• 6-sided CAVEs
• Head mounted displays (HMDs)
• BOOMs (HMD mounted on stand)
• 2) Partial immersive displays
• 3-5 sides CAVEs
• Panorama, Powerwalls
• Monitors

14
Creating interaction for VR

• Based on a 6-month project during 2000
• Dividing the concept of interaction
• Orientation
• Navigation
• Manipulation
• selection/picking
• systemcontrol

15
Orientation in VR

• being able to look around in a virtuel
  environment developing a sense of presence

• Available field of view
• Full immersive displays
• Display concepts has build-in support for
  orientation
• Partial immersive displays
• Display concepts has build-in limits for
  orientation

16
1) Joystick/trackball

• The user rotates in the virtual world by moving
  the joystick/trackball from side to side

• Advantages easy to use gets the job done
• Disdvantages having to use an extra device
• Joystick best for fast/continuous movements
• Trackball best for precise/absolute movements
• Trackball greater feeling of control

17
2) Headtracking (hypersensitive)

• The users rotation is enhanced, making it
  possible to view 360º within rotation of 160º

• Advantages
• No extra device, easy to use, makes it posible to
  do very fast movements (useful in eg. Quake)
• Disadvantages
• Hard to gain a feeling of pressence, easy to
  loose ones orientation in the virtuel world ?

18
3) Headtracking (with zones)

• When the user looks towards the edge of the
  screen the virtual world rotates in that direction

• Advantages
• No extra device, easy to use, preserves ones
  feeling of orientation
• Disadvantages
• Limits the freedom of movement creates a
  conflict between orientation in the physical and
  the virtual world

19
Navigation i VR

• being able to move/find ones way through a
  virtual world

• To a large extend a question of app. context
• Virtuel worlds vs. virtuel objects
• Visualizations/simulations vs. eg. art, games,
  etc.

20
1) Joystick with 6DOF tracking

• The user flyes in the direction he moves the
  joystick.

• Advantages
• Easy to learn, easy to use. Large freedom of
  movement
• Disadvantages
• The joystick is not suited for both fast and
  precise movements has a need of more gears
• the joystick is best suited for continuous
  movements

21
2) Trackball with 6DOF tracking

• The user flyes in the direction he moves the
  trackball

• Advantages
• Easy to learn, easy to use. Large freedom of
  movement
• Small movements greater feeling of control
• Disadvantages
• The trackball is not suited for flying long
  distances
• the trackball is best suited for absolute
  movements

22
3) Navigation zones

• The user is surrounded by a number of virtual
  zones. By pointing with his hand the user
  activates different navigation modes

• Advantages
• No extra devices, no need for buttonpresses
• Easy to combine with manipulation
• Disadvantages
• Limits freedom of movement
• Can be difficult to understand

23
Manipulation i VR

• Two primary approaches
• Simple virtual hand
• Ray-casting (virtuel laserbeam)

24
The virtual hand approach

• The users hand is represented in VR using
  positiontracking

• Advantages
• Very natural and easy to learn
• Easy to rotate and position objects
• Disadvantages
• Limited area of manipulation

25
The ray casting approach

• The user interacts using a virtual laserbeam
  pointing from the inputdevice or physical hand

• Advantages
• Objects at a distance can be selected/picked
• (also) very natural and easy to learn
• Disadvantages
• Ineffective for selecting small objects at a long
  distance
• Not good for rotating and positioning objects

26
An in between solution?

• The go-go hand (Poupyrev et al.) The virtual
  hand is not liniar, but lets user stretch his
  virtual arm for selection/manipulation of objects
  far away

• Advantages
• Allows manipulation in a large range of distances
• Disadvantages
• Range is still limited, interaction is not natural

27
1) Picking by touching

• The user selects objects for manipulation by
  touching them

• Advantages
• No need for buttons (works with eg.
  computervision)
• Disadvantages
• Question how to drop selected objects?
• Constrained objects (handles, levers etc.)
• Not constrained objects (torches, tools etc.)

28
2) Picking using buttonpresses

• The user selects objects for manipulation by
  touching them AND pressing a button

• Advantages
• Clear indication of intended interaction
• Easy to use (also for the ray casting approach)
• Disadvantages
• None but having to use a physical device

29
3) Using gloves

• The user selects objects for manipulation using
  gestures (eg. touching closing hand)

• Advantages
• Perhaps more natural ? (but is that always
  better?)
• Gloves are (can be) comfortable and easy to use
• Disadvantages
• No clear distinction between interacting in
  physical or virtual world

30
Combining interaction modes

• Optimize combinations
• Devices, display, context
• Implementation of devices
• Orientation
• Navitagion
• Manipulation

31
Conclusions

• There is NOT one optimal way to design
  interaction for virtual reality
• Optimal interaction techniques is dependant of
  the type of display used
• Solution not trivial
• Hard work
• Creativity
• Usability tests

32
Mulighederne for at udvikle interaktive VR
applikationer ved VRCN i dag...
33
Udvikling af VR-apps. ved VRCN

• VR området udgør stadig a new frontier
• Standarder er stort set ikke-eksisterende
• Mangel på tradition og konventioner i design
• Mangel på gode udviklingsværktøjer
• Hovedregel VR kræver programmering!
• Skal man lave noget fedt, må man have fingrene
  ned i skidtet!
• Men er dette reelt anderledes end ift. mm og
  web?

34
Udvikling af VR-apps. ved VRCN

• Division vs. CaveLib
• Division (grafisk interface)
• Relativt let at gå til
• Begrænset ift. programmeringsmuligheder
• CaveLib (/C bibliotek)
• Store muligheder for programmering af det fede...
• Svært at bruge for ikke-programmører

35
Interaktion i Division

• Baseret på en virtuel krop - metafor
• Brugeren repræsenteres ved en række relaterede
  kropsdele som torso, head, eyes, leftHand,
  rightHand osv.
• Den virtuelle krop kan konfigureres i en .bod
  tekstfil (tekst)
• Inputdevices sættes op og tilknyttes den
  virtuelle krop i en registry tekstfil
• En række standard body/registry filer kan
  anvendes og/eller tilpasses

36
Divisions tilgang til interaktion

• Fordele
• Let forståelig/overskuelig metafor
• Overskuelig at konfigurere/anvende
• Ulemper
• Begrænser hurtigt mulighederne for at få det som
  man vil have det
• Er begrænset ift. anvendelsen af andre typer
  devices end positionstrackere og joysticks
  (f.eks. handsker)

37
Understøttede devices

• Polhemus fastrak/isotrak (stylus, 3Ball, plain)
• Computervision i CAVE
• Wanda/TrackMan
• Spacemouse
• 2D mus/keyboard
• Handsker (som 5x knapper comming soon)
• Meget konfigurerbare nye drivere!!!
• p.t. ca. 45 forskellige registry kombinationer af
  disse devices til CAVE, Panorama og desktop
  setup

38
Videre arbejde med VR

• Kombinationer af orientering, navigation og
  manipulation i konkrete anvendelseskontekster
• Interaktion på tværs af arenaer (Orakel and
  beyond)
• Flerbruger interaktion
• Trådløs interaktion i CAVE fortsat fokus på
  usability, anvendelsesområder og forbedring af
  teknologi
• Anvendelse af spildevices til interaktion i VR
• Force-feedback v. interaktion i VR