0
13/03/14 pag. 1
Information visualization lecture 4
presentation
Katrien Verbert
Department of Computer Science
Faculty of...
13/03/14 pag. 2
Support	
  for	
  report	
  prepara+on.	
  Many	
  sources	
  of	
  content	
  are	
  visible	
  and	
  re...
13/03/14 pag. 3
The presentation issue
present (tr.v): to offer to view; display.
13/03/14 pag. 4
overview
Space	
  limita+ons	
  
	
  
•  Scrolling	
  
•  Overview	
  +	
  detail	
  
•  Distor+on	
  
•  ...
13/03/14 pag. 5
Space limitations
13/03/14 pag. 6
7.1 A PROBLEM
Many of us have found ourselves with a
report that has to be completed by a
deadline, with t...
13/03/14 pag. 7
Overview + detail
13/03/14 pag. 8
Source:	
  Courtesy	
  Colin	
  Grimshaw	
  
Overview + detail
13/03/14 pag. 9
Overview + detail
hDp://www.datavis.ca/milestones/	
  
	
  
13/03/14 pag. 10
A	
  journey	
  north	
  towards	
  Halifax	
  requires	
  detail	
  of	
  the	
  town	
  (Huddersfield)	
...
13/03/14 pag. 11
The	
  use	
  of	
  a	
  real	
  or	
  digitally	
  simulated	
  magnifying	
  glass	
  masks	
  detail	
...
13/03/14 pag. 12
The	
  DragMag	
  technique	
  allows	
  flexible	
  posi+oning	
  of	
  the	
  region	
  to	
  be	
  magn...
13/03/14 pag. 13
Connection between the detail and overview presentations missing
Overview + detail
Issues?	
  
13/03/14 pag. 14
Focus + context
13/03/14 pag. 15
Metaphor	
  illustra+ng	
  the	
  
principle	
  of	
  the	
  Bifocal	
  Display	
  
(a) An information sp...
13/03/14 pag. 16
An early illustration of the bifocal display principle
13/03/14 pag. 17
An early illustration of the bifocal display principle
13/03/14 pag. 18
Bifocal display features
1.  Distortion: available display area is allocated to two different regions
–  ...
13/03/14 pag. 19
What is the Bifocal Display Doing?
Transforming the information
space to the display space
7.19	
  
Infor...
13/03/14 pag. 20
A	
  sequence	
  of	
  amino	
  acids	
  within	
  a	
  protein	
  
Source:	
  Courtesy	
  of	
  Tom	
  O...
13/03/14 pag. 21
Table lens without distortion
13/03/14 pag. 22
Table lens with distortion
13/03/14 pag. 23
Table Lens: demo
hDp://www.youtube.com/watch?v=qWqTrRAC52U	
  
	
  
13/03/14 pag. 24
Schematic representation of X-distortion
13/03/14 pag. 25
Schematic representation of combined X- and Y-
distortion
13/03/14 pag. 26
hDp://www.youtube.com/watch?v=D1ediZXIDkc	
  
	
  
13/03/14 pag. 27
Distorted presentation of the London
Underground map
13/03/14 pag. 28
11Sun
12 Mon
13 Tue
14 Wed
15 Thur
16 Fri
17Sat
Fly LA
Kathy to airport Model Maker
Check slides, notes.
...
13/03/14 pag. 29
Visual	
  designer’s	
  sketch	
  of	
  the	
  applica+on	
  of	
  the	
  flip-­‐zoom	
  technique	
  to	
...
13/03/14 pag. 30
Source:	
  Courtesy	
  David	
  Baar,	
  IDELIX	
  SoFware	
  Inc.	
  
Distorted map on a PDA, showing th...
13/03/14 pag. 31
Source:	
  Courtesy	
  IDELIX	
  and	
  Mitsubishi	
  
Distorted map on a table
13/03/14 pag. 32
Equal X- and Y-distortion centred around a manually
chosen location in the Macintosh OSX ‘dock’
13/03/14 pag. 33
The Perspective Wall applies a 3D effect to the
bifocal display
13/03/14 pag. 34
Advantages Perspective Wall
•  User can adjust ratio of detail to context
•  Smooth animation helps user ...
13/03/14 pag. 35
Perspective Wall
Perspective gives smoother transition from focus to context
Informa+on	
  
space	
  
Dis...
13/03/14 pag. 36
overview
Space	
  limita+ons	
  
	
  
•  Scrolling	
  
•  Overview	
  +	
  detail	
  
•  Distor+on	
  
• ...
13/03/14 pag. 37
Suppression
•  Applies a distance function
and relevance function
•  Less relevant other items are
droppe...
13/03/14 pag. 38
Suppression
•  Originally proposed by Furnas (1986), but many variations of
applications.
•  Basic idea: ...
13/03/14 pag. 39
Degree of interest (DOI) function:
DOI(a|.=b)	
  =	
  API(a)	
  –	
  D(A,b)	
  
•  DOI(a|.=b):	
  DOI	
  ...
13/03/14 pag. 40
G
P
President
S
M N
F
K
The organization tree of a company
13/03/14 pag. 41
1
2
3 3
4 4
22
1
1
1
22
P
Focus
Showing the ‘distance’ of each node from the
focus of attention
13/03/14 pag. 42
Focus
Context
P
S
M NK
The context defined by setting an upper threshold
of unity for distance from a foc...
13/03/14 pag. 43
Example of a display that might be associated
with the focus and context
13/03/14 pag. 44
Each	
  node	
  in	
  the	
  organiza+on	
  tree	
  has	
  been	
  assigned	
  an	
  a	
  priori	
  impor...
13/03/14 pag. 45
Degree of Interest (DoI)
DoI = API – D
Expressed as a function of two quantities:
•  A priori importance ...
13/03/14 pag. 46
Segng	
  a	
  lower	
  limit	
  of	
  6	
  for	
  DoI	
  iden+fies	
  the	
  nodes	
  within	
  the	
  sha...
13/03/14 pag. 47
Part	
  of	
  an	
  
engineering	
  drawing	
  
Applications of DoI concept
13/03/14 pag. 48
The	
  engineering	
  drawing	
  
simplified	
  in	
  the	
  context	
  
of	
  a	
  suspected	
  fault	
  ...
13/03/14 pag. 49
Illustra+ng	
  the	
  concept	
  of	
  a	
  magic	
  lens.	
  (a)	
  shows	
  a	
  conven+onal	
  map	
  ...
13/03/14 pag. 50
hDp://www.youtube.com/watch?v=2bYDKbzocSg	
  
	
  
13/03/14 pag. 51
A	
  molecular	
  surface	
  of	
  the	
  protein	
  transferase	
  coloured	
  by	
  electrosta+c	
  pot...
13/03/14 pag. 52
A 3D Flexible and Tangible Magic Lens in
Augmented Reality
www.youtube.com/watch?v=PKegByAZ0kM	
  
13/03/14 pag. 53
	
  
	
  
A	
  combina+on	
  of	
  rubber-­‐sheet	
  distor+on	
  and	
  suppression	
  lead	
  to	
  a	
...
13/03/14 pag. 54
The rubber-sheet distortion technique employed
in the map
13/03/14 pag. 55
Historical note
•  Distortion and suppression appeared in early 1980s
•  Need to maintain a balanced view...
13/03/14 pag. 56
Fisheye view
•  Farrand also coined the term “fisheye”
•  Nowadays appears to refer to both distortion an...
13/03/14 pag. 57
Fisheye Menus
•  Here is the same idea applied to menus
–  Ben	
  Bederson,	
  University	
  of	
  Maryla...
13/03/14 pag. 58
Fisheye View, Polyfocal Display
Can	
  distort	
  boundaries	
  because	
  applied	
  radially	
  rather	...
13/03/14 pag. 59
hDp://www.cs.umd.edu/class/fall2002/
cmsc838s/+chi/fisheye.html	
  
	
  
13/03/14 pag. 60
hDp://www.youtube.com/watch?v=dEowKzbDpKU	
  
hDps://plus.google.com/+MiguelRodriguez/posts/YqBm18xQgQc	
  
	
  
13/03/14 pag. 62
Source:	
  By	
  kind	
  permission	
  of	
  Patrick	
  Baudisch	
  
The use of representation (by a ‘hal...
13/03/14 pag. 63
overview
Space	
  limita+ons	
  
	
  
•  Scrolling	
  
•  Overview	
  +	
  detail	
  
•  Distor+on	
  
• ...
13/03/14 pag. 64
Panning	
  is	
  the	
  smooth	
  movement	
  of	
  a	
  viewing	
  frame	
  over	
  a	
  2D	
  image	
  ...
13/03/14 pag. 65
	
  
Zooming	
  is	
  the	
  increasing	
  magnifica+on	
  of	
  a	
  decreasing	
  frac+on	
  of	
  an	
 ...
13/03/14 pag. 66
Zooming
•  Conventional zooming-in
–  No	
  change	
  in	
  data	
  or	
  representa+on	
  –	
  only	
  fi...
13/03/14 pag. 67
A space-scale diagram relevant
to combined zooming and
panning
Furnas	
  and	
  Bederson	
  (1995)	
  
13/03/14 pag. 68
Google earth
13/03/14 pag. 69
Exploring Information Spaces by Using
Tangible Magic Lenses
hDp://www.youtube.com/watch?v=h-­‐mF4_OAhU0	
...
13/03/14 pag. 70
Semantic zoom
•  Previous example: geometric zoom
–  Con+nuous	
  
–  Zooming-­‐in:	
  filtering	
  and	
 ...
13/03/14 pag. 71
A combination of geometric and semantic zoom
13/03/14 pag. 72
overview
Space	
  limita+ons	
  
	
  
•  Scrolling	
  
•  Overview	
  +	
  detail	
  
•  Distor+on	
  
• ...
13/03/14 pag. 73
A	
  collec+on	
  of	
  images	
  is	
  presented,	
  one	
  at	
  a	
  +me,	
  at	
  a	
  rapid	
  rate	...
13/03/14 pag. 74
Tile mode: concurrent presentation of images
opposed to ‘slide show mode’
13/03/14 pag. 75
‘Floa+ng	
  RSVP’	
  in	
  which	
  
images	
  appear	
  to	
  approach	
  
the	
  viewer	
  from	
  a	
 ...
13/03/14 pag. 76
The	
  contents	
  of	
  an	
  online	
  bookstore	
  are	
  presented	
  in	
  ‘collage	
  mode’	
  RSVP...
13/03/14 pag. 77
An	
  interface	
  facilita+ng	
  the	
  browsing	
  of	
  posters	
  adver+sing	
  videos.	
  	
  Cursor...
13/03/14 pag. 78
Space-time trade-off
13/03/14 pag. 79
Space-time trade-off
13/03/14 pag. 80
	
  
An	
  experiment	
  to	
  test	
  a	
  subject’s	
  ability	
  to	
  recognise	
  the	
  presence	
 ...
13/03/14 pag. 81
Representa+on	
  of	
  limits	
  on	
  display	
  area	
  and	
  total	
  presenta+on	
  +me	
  by	
  a	
...
13/03/14 pag. 82
Source:	
  Courtesy	
  of	
  Katy	
  Cooper	
  
Three ‘static’ image presentation modes (A, B, C) and
thr...
13/03/14 pag. 83
Slide show
13/03/14 pag. 84
Mixed
13/03/14 pag. 85
Tile
13/03/14 pag. 86
Diagonal
13/03/14 pag. 87
Ring
13/03/14 pag. 88
Stream
13/03/14 pag. 89
Source:	
  Courtesy	
  of	
  Katy	
  Cooper	
  
Favorite mode?
13/03/14 pag. 90
The	
  accuracy	
  with	
  which	
  the	
  presence	
  or	
  absence	
  of	
  a	
  target	
  image	
  was...
13/03/14 pag. 91
The	
  (sta+c)	
  slide-­‐show,	
  mixed	
  and	
  +le	
  image	
  presenta+on	
  modes	
  account	
  for...
13/03/14 pag. 92
Almost	
  all	
  the	
  least	
  preferred	
  image	
  presenta+on	
  modes	
  were	
  moving	
  modes	
 ...
13/03/14 pag. 93
	
  
A	
  simple	
  representa+on	
  of	
  eye-­‐gaze	
  behaviour.	
  The	
  rapid	
  saccades	
  are	
 ...
13/03/14 pag. 94
Eye-gaze trajectory slide show
13/03/14 pag. 95
Eye-gaze trajectory tile mode
13/03/14 pag. 96
Eye-gaze trajectories mixed mode
13/03/14 pag. 97
Eye-gaze trajectories diagonal mode for
subject who disliked the mode
13/03/14 pag. 98
Eye-gaze trajectories diagonal mode for
subject who liked the mode
13/03/14 pag. 99
Eye-gaze trajectory stream for subject
who disliked the mode
13/03/14 pag. 100
The	
  acquisi+on	
  of	
  an	
  expanding	
  target.	
  (a)	
  The	
  dormant	
  appearance	
  of	
  th...
13/03/14 pag. 101
An	
  experiment	
  in	
  which	
  a	
  subject	
  first	
  views	
  the	
  rapid	
  (e.g.,	
  10	
  per	...
13/03/14 pag. 102
An	
  experiment	
  in	
  which	
  a	
  collec+on	
  of	
  images	
  is	
  presented	
  to	
  a	
  subje...
13/03/14 pag. 103
A	
  third	
  palleDe	
  for	
  the	
  interac+on	
  designer,	
  addressing	
  issues	
  of	
  presenta...
13/03/14 pag. 104
Visual Information Seeking: Mantra
Overview, zoom & filter, details-on-demand
Overview, zoom & filter, d...
13/03/14 pag. 105
Shneiderman’s “7 Tasks”
•  Overview task
–  overview of entire collection
•  Zoom task
–  zoom in on ite...
13/03/14 pag. 106
Questions?
13/03/14 pag. 107
Readings
•  Chapter 4
13/03/14 pag. 108
References
•  Furnas, G. W., & Bederson, B. B. (1995, May). Space-scale
diagrams: Understanding multisca...
13/03/14 pag. 109
Research presentations
13/03/14 pag. 110
Research presentations
•  Schedule on PointCarré
•  Select a second paper in the same slot for questions...
13/03/14 pag. 111
Team project
13/03/14 pag. 112
Team project milestones
1.  Form teams
2.  Project proposal
3.  Intermediate presentation
4.  Final pres...
13/03/14 pag. 113
Project proposal
1 page description of your intended project:
–  mo+va+on	
  
–  which	
  datasets	
  yo...
13/03/14 pag. 114
Data collection
•  https://docs.google.com/forms/d/
1gHwVWHZLzWdSz1F37jA1Gungrl56bT215M6FYW3YqGY/
viewfo...
Upcoming SlideShare
Loading in...5
×

Information visualization: presentation

1,079

Published on

Published in: Education
0 Comments
1 Like
Statistics
Notes
  • Be the first to comment

No Downloads
Views
Total Views
1,079
On Slideshare
0
From Embeds
0
Number of Embeds
2
Actions
Shares
0
Downloads
33
Comments
0
Likes
1
Embeds 0
No embeds

No notes for slide

Transcript of "Information visualization: presentation"

  1. 1. 13/03/14 pag. 1 Information visualization lecture 4 presentation Katrien Verbert Department of Computer Science Faculty of Science Vrije Universiteit Brussel katrien.verbert@vub.ac.be
  2. 2. 13/03/14 pag. 2 Support  for  report  prepara+on.  Many  sources  of  content  are  visible  and  ready  to  hand   A problem
  3. 3. 13/03/14 pag. 3 The presentation issue present (tr.v): to offer to view; display.
  4. 4. 13/03/14 pag. 4 overview Space  limita+ons     •  Scrolling   •  Overview  +  detail   •  Distor+on   •  Suppression   •  Zoom  and  pan           Time  limita+ons     •  Rapid  serial  visual   presenta+on   •  Eye-­‐gaze                
  5. 5. 13/03/14 pag. 5 Space limitations
  6. 6. 13/03/14 pag. 6 7.1 A PROBLEM Many of us have found ourselves with a report that has to be completed by a deadline, with the result (Figure 7.1) that the dining room table, extended to its 12- guest state, is covered by piles of paper as well as reports, books, clippings and slides; perhaps with more arranged on the floor and on a couple of chairs. There may even be piles on top of piles. Such a presentation of vital information makes a lot of sense: everything relevant is to hand (hopefully!) and, moreover, its very visibility acts as a reminder (Bolt, 1984, page 2) of what might be relevant at any particular juncture, possibly triggering a situated action (Suchman, 1987). In this environment I can concentrate on creative tasks rather than organisation. Despite the availability of high-resolution displays and powerful workstations I still write most of my reports in this way. Why? Because the display area provided by the typical workstation is far too small to support, visibly, all the sources that are relevant to my composition. 7.2 THE PRESENTATION PROBLEM I am not alone in the sense of having too much data to fit onto a small screen. A very large and expensive screen, for example, would be needed to display the London Underground map in sufficient detail(Figure 1.1), and it would be difficult or impossible to present, on a normal display, the complete organisation chart of IBM or ICI. Moreover, the recent emergence of small and mobile information and communication devices such as PDAs and wearable displays has additionally identified a pressing need for a solution to the ‘ too much data, too little display area’ problem: the presentation problem. How can it be solved, mindful of the need to support the activity of visualising the underlying data? 7.2.1 Scrolling An obvious solution is to scroll the data into and out of the visible area. In other words, to provide a means whereby a long document can be moved past a window until it reaches the required ‘page’ (Figure 7.2). This mechanism is widely used, but carries with it many penalties. One relates to the "Where am I?" problem: I’m working on Chapter 2, (it may be section 2.3, I don’t know) and I want to remind myself of a figure that is in chapter 5, it may be in section 5.3 – or was it 5.6? All I can do is operate the scrolling mechanism and look out for the figure I need, albeit assisted by various cues such as the page number indicated in the scrolling mechanism. With a scrolling mechanism, most of a document is hidden from view. I have the same problem when using a microfilm reader, with the additional complication that if I move the tray to the left, the image moves to the right. A similar difficulty applies to my use of the famous London ‘AtoZ’ street directory. I’m driving along a road that goes off the edge of the page, so I desperately need whatever page contains the continuation of that road (and quickly!). Even if I get it, I will typically have trouble locating the same road on the new page. These and other similar problems can be ameliorated by the provision of context. Much of this chapter, in fact, is concerned with deciding how to provide context. Scrolling
  7. 7. 13/03/14 pag. 7 Overview + detail
  8. 8. 13/03/14 pag. 8 Source:  Courtesy  Colin  Grimshaw   Overview + detail
  9. 9. 13/03/14 pag. 9 Overview + detail hDp://www.datavis.ca/milestones/    
  10. 10. 13/03/14 pag. 10 A  journey  north  towards  Halifax  requires  detail  of  the  town  (Huddersfield)  through  which   the  traveller  passes   Overview + detail
  11. 11. 13/03/14 pag. 11 The  use  of  a  real  or  digitally  simulated  magnifying  glass  masks  detail  around  the  magnified  region     Overview + detail
  12. 12. 13/03/14 pag. 12 The  DragMag  technique  allows  flexible  posi+oning  of  the  region  to  be  magnified     Overview + detail
  13. 13. 13/03/14 pag. 13 Connection between the detail and overview presentations missing Overview + detail Issues?  
  14. 14. 13/03/14 pag. 14 Focus + context
  15. 15. 13/03/14 pag. 15 Metaphor  illustra+ng  the   principle  of  the  Bifocal  Display   (a) An information space containing documents, emails, etc. (b) The same space wrapped around two uprights. (c) Appearance of the information space when viewed from an appropriate direction direction of view Distortion
  16. 16. 13/03/14 pag. 16 An early illustration of the bifocal display principle
  17. 17. 13/03/14 pag. 17 An early illustration of the bifocal display principle
  18. 18. 13/03/14 pag. 18 Bifocal display features 1.  Distortion: available display area is allocated to two different regions –  Focus  (undistorted)   –  Context  (distorted)   2.  Information moves smoothly and continuously from context to focus 3.  Display affords for representation –  opportunity  to  use  two  dimensions   –  for  instance,  +me  assigned  to  horizontal  axis   –  type  of  item  to  Y-­‐axis   4.  Main purpose –  Focus:  provide  detail   –  Context:  awareness  and  iden6fica6on     5.  Manual control
  19. 19. 13/03/14 pag. 19 What is the Bifocal Display Doing? Transforming the information space to the display space 7.19   Informa+on   space   Display   Space   Normal display Informa+on   space   Display   Space   Bifocal   display   context   focus   Slide  source:  Ken  Brodlie  
  20. 20. 13/03/14 pag. 20 A  sequence  of  amino  acids  within  a  protein   Source:  Courtesy  of  Tom  Oldfield   Applications of distortion technique
  21. 21. 13/03/14 pag. 21 Table lens without distortion
  22. 22. 13/03/14 pag. 22 Table lens with distortion
  23. 23. 13/03/14 pag. 23 Table Lens: demo hDp://www.youtube.com/watch?v=qWqTrRAC52U    
  24. 24. 13/03/14 pag. 24 Schematic representation of X-distortion
  25. 25. 13/03/14 pag. 25 Schematic representation of combined X- and Y- distortion
  26. 26. 13/03/14 pag. 26 hDp://www.youtube.com/watch?v=D1ediZXIDkc    
  27. 27. 13/03/14 pag. 27 Distorted presentation of the London Underground map
  28. 28. 13/03/14 pag. 28 11Sun 12 Mon 13 Tue 14 Wed 15 Thur 16 Fri 17Sat Fly LA Kathy to airport Model Maker Check slides, notes. Family barbeque Fly LHR Kathy to collect Chapter 2/ see Dave March JulyJuneMayAprilMar Aug Sept Oct Flight to SFO Tutorial set-up Tutorial United flight Heathrow Pointer Color OHs Jane+John Call Kathy Combined X- and Y-distortion provides a convenient calendar interface
  29. 29. 13/03/14 pag. 29 Visual  designer’s  sketch  of  the  applica+on  of  the  flip-­‐zoom  technique  to  the  presenta+on  of   photographs  on  a  Nokia  mobile  phone   Source:  Courtesy  Ron  Bird  
  30. 30. 13/03/14 pag. 30 Source:  Courtesy  David  Baar,  IDELIX  SoFware  Inc.   Distorted map on a PDA, showing the continuity of transportation links
  31. 31. 13/03/14 pag. 31 Source:  Courtesy  IDELIX  and  Mitsubishi   Distorted map on a table
  32. 32. 13/03/14 pag. 32 Equal X- and Y-distortion centred around a manually chosen location in the Macintosh OSX ‘dock’
  33. 33. 13/03/14 pag. 33 The Perspective Wall applies a 3D effect to the bifocal display
  34. 34. 13/03/14 pag. 34 Advantages Perspective Wall •  User can adjust ratio of detail to context •  Smooth animation helps user perceive object constancy •  Relationship between detail and context is consistent: objects bend around the corner Slide  source:  Ken  Brodlie  
  35. 35. 13/03/14 pag. 35 Perspective Wall Perspective gives smoother transition from focus to context Informa+on   space   Display   Space   Perspective Wall context focus Slide  source:  Ken  Brodlie  
  36. 36. 13/03/14 pag. 36 overview Space  limita+ons     •  Scrolling   •  Overview  +  detail   •  Distor+on   •  Suppression   •  Zoom  and  pan           Time  limita+ons     •  Rapid  serial  visual   presenta+on   •  Eye-­‐gaze                
  37. 37. 13/03/14 pag. 37 Suppression •  Applies a distance function and relevance function •  Less relevant other items are dropped from the display •  Classic example: New Yorker’s idea of the world
  38. 38. 13/03/14 pag. 38 Suppression •  Originally proposed by Furnas (1986), but many variations of applications. •  Basic idea: more relevant information presented in great detail; the less relevant information presented as an abstraction. •  Relevance is computed on basis of the importance of information elements and their distance to the focus.
  39. 39. 13/03/14 pag. 39 Degree of interest (DOI) function: DOI(a|.=b)  =  API(a)  –  D(A,b)   •  DOI(a|.=b):  DOI  of  a,  given  the  current  focus  is  b.   •  API(a):  sta+c  global  a  priori  importance  measure.   •  D(a,b):  distance  between  a  and  b.  
  40. 40. 13/03/14 pag. 40 G P President S M N F K The organization tree of a company
  41. 41. 13/03/14 pag. 41 1 2 3 3 4 4 22 1 1 1 22 P Focus Showing the ‘distance’ of each node from the focus of attention
  42. 42. 13/03/14 pag. 42 Focus Context P S M NK The context defined by setting an upper threshold of unity for distance from a focus
  43. 43. 13/03/14 pag. 43 Example of a display that might be associated with the focus and context
  44. 44. 13/03/14 pag. 44 Each  node  in  the  organiza+on  tree  has  been  assigned  an  a  priori  importance  (API)       10 9 9 8 7 7 7 8 8 6 8 8 6 9 API
  45. 45. 13/03/14 pag. 45 Degree of Interest (DoI) DoI = API – D Expressed as a function of two quantities: •  A priori importance (API) •  Distance (D) between an item and the item currently in focus
  46. 46. 13/03/14 pag. 46 Segng  a  lower  limit  of  6  for  DoI  iden+fies  the  nodes  within  the  shaded  region   8 6 6 8 6 6 6 4 4 4 6 6 4 8 Focus Context Nodal values of degree of interest (=API – D)
  47. 47. 13/03/14 pag. 47 Part  of  an   engineering  drawing   Applications of DoI concept
  48. 48. 13/03/14 pag. 48 The  engineering  drawing   simplified  in  the  context   of  a  suspected  fault   Applications of DoI concept
  49. 49. 13/03/14 pag. 49 Illustra+ng  the  concept  of  a  magic  lens.  (a)  shows  a  conven+onal  map  of  an  area,  (b)  shows  the   loca+on  of  services  (gas,  water  and  electricity  pipes)  in  the  same  area,  and  (c)  a  (movable)  magic   lens  shows  services  in  an  area  of  interest,  in  context   Application in magic lens technique
  50. 50. 13/03/14 pag. 50 hDp://www.youtube.com/watch?v=2bYDKbzocSg    
  51. 51. 13/03/14 pag. 51 A  molecular  surface  of  the  protein  transferase  coloured  by  electrosta+c  poten+al  bound  to   DNA  shown  as  a  schema+c.  (ID  =  10mh).  The  magic  lens  window  allows  a  view  of  the   atomic  structure  bonding  to  be  shown,  with  the  bound  ligand  structure  highlighted  as   cylinders,  thereby  providing  a  view  inside  the  protein   Source:  By  kind  permission  of  Tom  Oldfield  and  Michael  Hartshorn   Magic lens
  52. 52. 13/03/14 pag. 52 A 3D Flexible and Tangible Magic Lens in Augmented Reality www.youtube.com/watch?v=PKegByAZ0kM  
  53. 53. 13/03/14 pag. 53     A  combina+on  of  rubber-­‐sheet  distor+on  and  suppression  lead  to  a  map  appropriate  to  a   journey  from  one  city  to  another   Combined distortion and suppression
  54. 54. 13/03/14 pag. 54 The rubber-sheet distortion technique employed in the map
  55. 55. 13/03/14 pag. 55 Historical note •  Distortion and suppression appeared in early 1980s •  Need to maintain a balanced view of focus + context identified earlier – for example by Farrand (1973) “an effective transformation must somehow maintain global awareness while providing detail” “… there is a need for presenting a display with 1. sufficient detail for interaction, while 2. maintaining global vision of the entire scene.”
  56. 56. 13/03/14 pag. 56 Fisheye view •  Farrand also coined the term “fisheye” •  Nowadays appears to refer to both distortion and suppression
  57. 57. 13/03/14 pag. 57 Fisheye Menus •  Here is the same idea applied to menus –  Ben  Bederson,  University  of  Maryland   •  See also: –  hDp://www.cs.umd.edu/hcil/fisheyemenu/fisheyemenu-­‐demo.shtml   ENV  2006  
  58. 58. 13/03/14 pag. 58 Fisheye View, Polyfocal Display Can  distort  boundaries  because  applied  radially  rather  than  x  y   1D  Fisheye   2D  Polyfocal   Slide  source:  Hornung  and  Zagreus    
  59. 59. 13/03/14 pag. 59 hDp://www.cs.umd.edu/class/fall2002/ cmsc838s/+chi/fisheye.html    
  60. 60. 13/03/14 pag. 60 hDp://www.youtube.com/watch?v=dEowKzbDpKU  
  61. 61. hDps://plus.google.com/+MiguelRodriguez/posts/YqBm18xQgQc    
  62. 62. 13/03/14 pag. 62 Source:  By  kind  permission  of  Patrick  Baudisch   The use of representation (by a ‘halo’) to provide context for a small display
  63. 63. 13/03/14 pag. 63 overview Space  limita+ons     •  Scrolling   •  Overview  +  detail   •  Distor+on   •  Suppression   •  Zoom  and  pan           Time  limita+ons     •  Rapid  serial  visual   presenta+on   •  Eye-­‐gaze                
  64. 64. 13/03/14 pag. 64 Panning  is  the  smooth  movement  of  a  viewing  frame  over  a  2D  image     Zoom and pan
  65. 65. 13/03/14 pag. 65   Zooming  is  the  increasing  magnifica+on  of  a  decreasing  frac+on  of  an  image  (or  vice  versa)   Zoom and pan
  66. 66. 13/03/14 pag. 66 Zooming •  Conventional zooming-in –  No  change  in  data  or  representa+on  –  only  filtering   –  Loss  of  context     •  ≠distortion whose purpose is to permit focusing rather than filtering •  Supports two cognitive tasks (Cairns and Craft 2005) –  Zooming-­‐in:  extraneous  informa+on  is  removed  from  visual  field  –  more   manageable  view   –  Zooming-­‐out:  reveals  hidden  informa+on  
  67. 67. 13/03/14 pag. 67 A space-scale diagram relevant to combined zooming and panning Furnas  and  Bederson  (1995)  
  68. 68. 13/03/14 pag. 68 Google earth
  69. 69. 13/03/14 pag. 69 Exploring Information Spaces by Using Tangible Magic Lenses hDp://www.youtube.com/watch?v=h-­‐mF4_OAhU0    
  70. 70. 13/03/14 pag. 70 Semantic zoom •  Previous example: geometric zoom –  Con+nuous   –  Zooming-­‐in:  filtering  and  loss  of  context   •  Semantic zoom –  Discrete  transi+on   –  Addi+onal  detail  
  71. 71. 13/03/14 pag. 71 A combination of geometric and semantic zoom
  72. 72. 13/03/14 pag. 72 overview Space  limita+ons     •  Scrolling   •  Overview  +  detail   •  Distor+on   •  Suppression   •  Zoom  and  pan           Time  limita+ons     •  Rapid  serial  visual   presenta+on   •  Eye-­‐gaze                
  73. 73. 13/03/14 pag. 73 A  collec+on  of  images  is  presented,  one  at  a  +me,  at  a  rapid  rate  (e.g.,  ten  per  second)     time Rapid serial visual presentation
  74. 74. 13/03/14 pag. 74 Tile mode: concurrent presentation of images opposed to ‘slide show mode’
  75. 75. 13/03/14 pag. 75 ‘Floa+ng  RSVP’  in  which   images  appear  to  approach   the  viewer  from  a  distance.   Sensi+ve  arrows  allow  the   speed  and  direc+on  of   ‘movement’  to  be  controlled   by  a  user   Source:  Courtesy  Kent  WiNenburg   Floating RSVP
  76. 76. 13/03/14 pag. 76 The  contents  of  an  online  bookstore  are  presented  in  ‘collage  mode’  RSVP,  simula+ng  the   placing  of  book  covers  on  a  table  in  sequence.  The  set  of  arrows  just  under  the  presenta+on   allows  control  of  the  speed  and  direc+on  of  presenta+on   Source:  Courtesy  Kent  WiNenburg   Collage mode RSVP
  77. 77. 13/03/14 pag. 77 An  interface  facilita+ng  the  browsing  of  posters  adver+sing  videos.    Cursor  movement  along  the   stacks  causes  posters  to  briefly  ‘pop  out’  sideways,  and  the  whole  bifocal  structure  can  be   scrolled  to  bring  a  video  of  interest  to  the  central  region,  where  a  mouse  click  will  cause  a  clip   from  a  video  to  be  played  (Lam  and  Pence  1997)   RSVP + bifocal principle
  78. 78. 13/03/14 pag. 78 Space-time trade-off
  79. 79. 13/03/14 pag. 79 Space-time trade-off
  80. 80. 13/03/14 pag. 80   An  experiment  to  test  a  subject’s  ability  to  recognise  the  presence  or  absence  of  a  previously   viewed  target  image  within  a  collec+on  presented  sequen+ally  at  a  rate  of  around  ten  per  sec.     Prior instruction to subject Subjectsʼ performance “Here is a target image. Tell me if this image appears in the sequence of N images youʼre about to see” Recognition about 80% to 90% successful time about 100 ms unrelated images Presentation of images Briefly glimpsed images
  81. 81. 13/03/14 pag. 81 Representa+on  of  limits  on  display  area  and  total  presenta+on  +me  by  a  ‘resource  box’     Display area Presentation time Space and time resources
  82. 82. 13/03/14 pag. 82 Source:  Courtesy  of  Katy  Cooper   Three ‘static’ image presentation modes (A, B, C) and three ‘moving’ image presentation modes (D, E, F)
  83. 83. 13/03/14 pag. 83 Slide show
  84. 84. 13/03/14 pag. 84 Mixed
  85. 85. 13/03/14 pag. 85 Tile
  86. 86. 13/03/14 pag. 86 Diagonal
  87. 87. 13/03/14 pag. 87 Ring
  88. 88. 13/03/14 pag. 88 Stream
  89. 89. 13/03/14 pag. 89 Source:  Courtesy  of  Katy  Cooper   Favorite mode?
  90. 90. 13/03/14 pag. 90 The  accuracy  with  which  the  presence  or  absence  of  a  target  image  was  reported  for  the   six  presenta+on  modes,  averaged  over  all  tasks  and  presenta+on  +mes.         1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 Slide-show Mixed Tile Diagonal Ring Stream Recognition accuracy Presentation modes
  91. 91. 13/03/14 pag. 91 The  (sta+c)  slide-­‐show,  mixed  and  +le  image  presenta+on  modes  account  for  three-­‐quarters  of   the  preferred  modes  (Cooper  et  al.  2006)  
  92. 92. 13/03/14 pag. 92 Almost  all  the  least  preferred  image  presenta+on  modes  were  moving  modes  and  the  stream   mode  accounted  for  over  half    
  93. 93. 13/03/14 pag. 93   A  simple  representa+on  of  eye-­‐gaze  behaviour.  The  rapid  saccades  are  shown  green,  the   fixa+ons  (F)  of  varying  dura+on  by  circles  of  propor+onate  size   F F F F F F F F Eye-gaze
  94. 94. 13/03/14 pag. 94 Eye-gaze trajectory slide show
  95. 95. 13/03/14 pag. 95 Eye-gaze trajectory tile mode
  96. 96. 13/03/14 pag. 96 Eye-gaze trajectories mixed mode
  97. 97. 13/03/14 pag. 97 Eye-gaze trajectories diagonal mode for subject who disliked the mode
  98. 98. 13/03/14 pag. 98 Eye-gaze trajectories diagonal mode for subject who liked the mode
  99. 99. 13/03/14 pag. 99 Eye-gaze trajectory stream for subject who disliked the mode
  100. 100. 13/03/14 pag. 100 The  acquisi+on  of  an  expanding  target.  (a)  The  dormant  appearance  of  the  image  collec+on,   and  (b)  its  appearance  when  the  cursor  rests  over  image  6     1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 10 11 12 13 14 15 16 17 18 19 209861 2 3 4 (a) (b) Manual control: ‘expanding target’ presentation mode
  101. 101. 13/03/14 pag. 101 An  experiment  in  which  a  subject  first  views  the  rapid  (e.g.,  10  per  second)  presenta+on  of  a   collec+on  of  images,  is  then  shown  a  single  image  and  asked  to  say  whether  that  image  was   part  of  the  collec+on.  Iden+fica+on  success  is  highly  dependent  upon  the  +me  elapsing   between  the  end  of  the  presenta+on  and  the  ques+oning  of  the  subject   Prior instruction to subject Presentation of image collection Subject’s performance about 100ms unrelated images time None The subject was shown an image and then asked, ‘Was this image present in the sequence you have just seen?’ Recognition success was 10% to 20% unless the question was aksed within about 4 seconds of the end of the presentation Models of human visual performance
  102. 102. 13/03/14 pag. 102 An  experiment  in  which  a  collec+on  of  images  is  presented  to  a  subject.  Each  image  is  presented   briefly  (e.g.,  for  100ms)  and  followed  by  a  ‘visual  mask’  las+ng  about  300ms.  Subjects  were  able   to  say,  with  a  considerable  degree  of  success,  whether  an  image  shown  arerwards  had  been   part  of  the  presenta+on   Prior  instruc+on     to  subject   Presenta+on  of  image  collec+on   Subject’s  performance   about  300ms   unrelated  images   +me   None   The  subject  was  shown  an  image  and  then     asked,  ‘Was  this  image  present  in  the     sequence  you  have  just  seen?’   Up  to  92%  recogni+on  success   etc  .  .  .  .  Visual   mask    .     Visual   mask       Visual   mask   about  100ms   Models of human visual performance
  103. 103. 13/03/14 pag. 103 A  third  palleDe  for  the  interac+on  designer,  addressing  issues  of  presenta+on   Presentation   concepts and   techniques   Scrolling   Overview+detail   Distortion   Suppression   Zoom   Pan   RSVP   Eye gaze   Recap
  104. 104. 13/03/14 pag. 104 Visual Information Seeking: Mantra Overview, zoom & filter, details-on-demand Overview, zoom & filter, details-on-demand Overview, zoom & filter, details-on-demand Overview, zoom & filter, details-on-demand Overview, zoom & filter, details-on-demand Overview, zoom & filter, details-on-demand Overview, zoom & filter, details-on-demand Overview, zoom & filter, details-on-demand Overview, zoom & filter, details-on-demand Overview, zoom & filter, details-on-demand Ben  Shneiderman,  1996  
  105. 105. 13/03/14 pag. 105 Shneiderman’s “7 Tasks” •  Overview task –  overview of entire collection •  Zoom task –  zoom in on items of interest •  Filter task –  filter out uninteresting items •  Details-on-demand task –  select an item or group to get details •  Relate  task   –  relate  items  or  groups  within  the   collec+on   •  History  task     –  keep  a  history  of  ac+ons  to  support   undo,  replay,  and  progressive   refinement   •  Extract  task   –  allow  extrac+on  of  sub-­‐collec+ons  and   of  the  query  parameters  
  106. 106. 13/03/14 pag. 106 Questions?
  107. 107. 13/03/14 pag. 107 Readings •  Chapter 4
  108. 108. 13/03/14 pag. 108 References •  Furnas, G. W., & Bederson, B. B. (1995, May). Space-scale diagrams: Understanding multiscale interfaces. In Proceedings of the SIGCHI conference on Human factors in computing systems (pp. 234-241). ACM Press/Addison- Wesley Publishing Co.. •  Shneiderman, B. (1996, September). The eyes have it: A task by data type taxonomy for information visualizations. In Visual Languages, 1996. Proceedings., IEEE Symposium on (pp. 336-343). IEEE. •  Some relevant notes: http://jcsites.juniata.edu/faculty/rhodes/ida/ presentation.html
  109. 109. 13/03/14 pag. 109 Research presentations
  110. 110. 13/03/14 pag. 110 Research presentations •  Schedule on PointCarré •  Select a second paper in the same slot for questions: e.g. session 1: http://doodle.com/rmpc9g8u3p2qzsy4 •  Links to doodle polls for all six sessions will be included in the schedule.
  111. 111. 13/03/14 pag. 111 Team project
  112. 112. 13/03/14 pag. 112 Team project milestones 1.  Form teams 2.  Project proposal 3.  Intermediate presentation 4.  Final presentation 5.  Short report due  27  Feb.   due  13  March   due  3  April   22  May   due  29  May  
  113. 113. 13/03/14 pag. 113 Project proposal 1 page description of your intended project: –  mo+va+on   –  which  datasets  you  will  use   –  current  status.  If  available,  first  designs.   –  problems/ques+ons   due 13 March If you want earlier feedback, send us your proposal earlier ;-)
  114. 114. 13/03/14 pag. 114 Data collection •  https://docs.google.com/forms/d/ 1gHwVWHZLzWdSz1F37jA1Gungrl56bT215M6FYW3YqGY/ viewform Or •  bit.ly/N6JTyD Anonymous! Choose your own ID. •  Please report your data ;-)
  1. A particular slide catching your eye?

    Clipping is a handy way to collect important slides you want to go back to later.

×