iat 814 introduction to visual analytics
DESCRIPTION
IAT 814 Introduction to Visual Analytics. Perception. Perceptual Processing. Seek to better understand visual perception and visual information processing Multiple theories or models exist Need to understand physiology and cognitive psychology. A Simple Model. Two stage process - PowerPoint PPT PresentationTRANSCRIPT
IAT 814 1
IAT 814 Introduction to Visual Analytics
Perception
Sep 11, 2013
IAT 814 2
Perceptual Processingg Seek to better understand visual
perception and visual information processing– Multiple theories or models exist– Need to understand physiology and
cognitive psychology
Sep 11, 2013
IAT 814 3
A Simple Modelg Two stage process
– Parallel extraction of low-level properties of scene
– Sequential goal-directed processing
Sep 11, 2013
EyeStage 1Early, parallel detection of color, texture, shape, spatial attributes
Stage 2Serial processing of object identification (using memory) and spatial layout, action
IAT 814 4
Stage 1 - Low-level, Parallel
g Neurons in eye & brain responsible for different kinds of information– Orientation, color, texture, movement, etc.– Arrays of neurons work in parallel– Occurs “automatically”– Rapid– Information is transitory, briefly held in iconic
store– Bottom-up data-driven model of processing– Often called “pre-attentive” processing
Sep 11, 2013
IAT 814 5
Stage 2 - Sequential, Goal-Directed
g Splits into subsystems for object recognition and for interacting with environment
g Increasing evidence supports independence of systems for symbolic object manipulation and for locomotion & action
g First subsystem then interfaces to verbal linguistic portion of brain, second interfaces to motor systems that control muscle movementsSep 11, 2013
IAT 814 6
Stage 2 Attributesg Slow serial processingg Involves working and long-term
memoryg Top-down processing
Sep 11, 2013
IAT 814 7
Preattentive Processingg How does human visual system
analyze images?– Some things seem to be done
preattentively, without the need for focused attention
– Generally less than 200-250 msecs (eye movements take 200 msecs)
– Seems to be done in parallel by low-level vision system
Sep 11, 2013
IAT 814 8
How Many 3’s?1281768756138976546984506985604982826762980985845822450985645894509845098094358590910302099059595957725646750506789045678845789809821677654876364908560912949686
Sep 11, 2013
IAT 814 9
How Many 3’s?1281768756138976546984506985604982826762980985845822450985645894509845098094358590910302099059595957725646750506789045678845789809821677654876364908560912949686
Sep 11, 2013
IAT 814 10
What Kinds of Tasks?g Target detection
– Is something there?g Boundary detection
– Can the elements be grouped?g Counting
– How many elements of a certain type are present?
Sep 11, 2013
IAT 814 11
Examples:g Where is the red circle? Left or right?g Put your hand up as soon as you see
it.
Sep 11, 2013
IAT 814 12
Pre-attentive Hue
g Can be done rapidly
Sep 11, 2013
IAT 814 13
Examples:g Where is the red circle? Left or right?g Put your hand up as soon as you see
it.
Sep 11, 2013
IAT 814 14
Shape
Sep 11, 2013
IAT 814 15
Examples:g Where is the red circle? Left or right?g Put your hand up as soon as you see
it.
Sep 11, 2013
IAT 814 16
Hue and Shape
g Cannot be done preattentivelyg Must perform a sequential searchg Conjuction of features (shape and hue)
causes itSep 11, 2013
IAT 814 17
Examplesg Is there a boundary:
– A connected chain of features that cross the rectangle
g Put your hand up as soon as you see it.
Sep 11, 2013
IAT 814 18
Fill and Shape
g Left can be done preattentively since each group contains one unique feature
g Right cannot (there is a boundary!) since the two features are mixed (fill and shape)
Sep 11, 2013
IAT 814 19
Examplesg Is there a boundary?
Sep 11, 2013
IAT 814 20
Hue versus Shape
g Left: Boundary detected preattentively based on hue regardless of shape
g Right: Cannot do mixed color shapes preattentively
Sep 11, 2013
IAT 814 21
Hue vs. Brightness
g Left: Varying brightness seems to interfereg Right: Boundary based on brightness can
be done preattentively
Sep 11, 2013
IAT 814 22
Preattentive Featuresg Certain visual forms lend themselves
to preattentive processingg Variety of forms seem to work
g In the next slide, spot the region of different shapes, both left and right
Sep 11, 2013
IAT 814 23
3-D Figures
g 3-D visual reality has an influence
Sep 11, 2013
IAT 814 24
Emergent Features
Sep 11, 2013
IAT 814 25
Potential PA Featuresg lengthg widthg sizeg curvatureg numberg terminatorsg intersectiong closure
g hueg intensityg flickerg direction of motiong stereoscopic depthg 3-D depth cuesg lighting direction
Sep 11, 2013
IAT 814 26
Key Perceptual Propertiesg Brightnessg Colorg Textureg Shape
Sep 11, 2013
IAT 814 27
Luminance/Brightnessg Luminance
– Measured amount of light coming from some place
g Brightness– Perceived amount of light coming
from source
Sep 11, 2013
IAT 814 28
Brightnessg Perceived brightness is non-linear
function of amount of light emitted by sourceTypically a power functionS = aIn
S - sensationI - intensity
g Very different on screen versus paper
Sep 11, 2013
IAT 814 29
Greyscaleg Probably not best way to encode
data because of contrast issues– Surface orientation and surroundings
matter a great deal– Luminance channel of visual system is
so fundamental to so much of perception• We can get by without color discrimination,
but not luminance
Sep 11, 2013
IAT 814 30
Greyscaleg White and Black are not fixed
Sep 11, 2013
IAT 814 31
Greyscaleg White and Black are not fixed!
Sep 11, 2013
IAT 814 32
Color Systemsg HSV: Hue,
Saturation, Value– Hue: Color type– Saturation:
“Purity” of color– Value: Brightness
Sep 11, 2013
IAT 814 33
CIE Spaceg The perceivable set of colors
Sep 11, 2013
IAT 814 34
CIE L*a*b*
g http://www.gamutvision.com/docs/printest.htmlSep 11, 2013
IAT 814 35
Color Categoriesg Are there certain
canonical colors?– Post & Greene ’86
had people name different colors on a monitor
– Pictured are ones with > 75% commonality
Sep 11, 2013
IAT 814 36
Luminanceg Foreground must be distinct from
background!
Sep 11, 2013
Can you read this text?Can you read this text?Can you read this text?Can you read this text?Can you read this text?Can you read this text?
IAT 814 37
Color for Categoriesg Can different colors be used for
categorical variables?– Yes (with care)– Colin Ware’s suggestion: 12 colors
• red, green, yellow, blue, black, white, pink, cyan, gray, orange, brown, purple
Sep 11, 2013
IAT 814 38
Why 12 colors?
Sep 11, 2013
IAT 814 39
Just-Noticeable Differenceg Which is brighter?
Sep 11, 2013
IAT 814 40
Just-Noticeable Differenceg Which is brighter?
Sep 11, 2013
(130, 130, 130) (140, 140, 140)
IAT 814 41
Weber’s Lawg In the 1830’s, Weber made measurements
of the just-noticeable differences (JNDs) in the perception of weight and other sensations.
g He found that for a range of stimuli, the ratio of the JND ΔS to the initial stimulus S was relatively constant:
ΔS / S = kSep 11, 2013
IAT 814 42
Weber’s Law
g Ratios more important than magnitude in stimulus detection
g For example: we detect the presence of a change from 100 cm to 101 cm with the same probability as we detect the presence of a change from 1 to 1.01 cm, even though the discrepancy is 1 cm in the first case and only .01 cm in the second.
Sep 11, 2013
IAT 814 43
Weber’s Lawg Most continuous variations in
magnitude are perceived as discrete steps
g Examples: contour maps, font sizesSep 11, 2013
IAT 814 44
Weber’s Lawg Most continuous variations in
magnitude are perceived as discrete steps
g Examples: contour maps, font sizesSep 11, 2013
IAT 814 45
Stevens’ Power Lawg Compare area of circles:
Sep 11, 2013
IAT 814 46
Stevens’ Power Law
s(x) = axb
s is the sensationx is the intensity of the
attributea is a multiplicative constantb is the power
b > 1: overestimateb < 1: underestimate
Sep 11, 2013
[graph from Wilkinson 99]
IAT 814 47
Stevens’ Power LawSensation Exponent Brightness 0.33 Smell 0.55 (Coffee) Loudness 0.6 Temperature 1.0 (Cold) Taste 1.3 (Salt) Heaviness 1.45 Electric Shock 3.5
Sep 11, 2013
[Stevens 1961]
IAT 814 48
Stevens’ Power Law
Experimental results for b:
Length .9 to 1.1Area .6 to .9Volume .5 to .8
Heuristic: b ~ 1/sqrt(dimensionality)
Sep 11, 2013
IAT 814 49
Stevens’ Power Lawg Apparent magnitude scaling
Sep 11, 2013
[Cartography: Thematic Map Design, p. 170, Dent, 96]
S = 0.98A0.87
[J. J. Flannery, The relative effectiveness of some graduated point symbols in the presentation of quantitative data, Canadian Geographer, 8(2), pp. 96-109, 1971] [slide from Pat Hanrahan]
IAT 814 50
Relative Magnitude Estimation
Most accurate
Least accurate
Position (common) scalePosition (non-aligned)
scale
LengthSlopeAngle
AreaVolumeColor (hue/saturation/value)
Sep 11, 2013
IAT 814 51
Color Sequenceg Can you order these?
Sep 11, 2013
IAT 814 52
Possible Color Sequences
Sep 11, 2013
Grey Scale
Full Spectral Scale
Partial Spectral Scale
Single Hue Scale
Double-ended Hue Scale
IAT 814 53
HeatMap
Sep 11, 2013
IAT 814 54
ColorBrewer
Sep 11, 2013
IAT 814 55
Color Purposesg Call attention to specific datag Increase appeal, memorabilityg Increase number of dimensions for
encoding data– Example, Ware and Beatty ‘88
• x,y - variables 1 & 2• amount of r,g,b - variables 3, 4, & 5
Sep 11, 2013
IAT 814 56
Using Colorg Modesty! Less is moreg Use blue in large regions, not thin
linesg Use red and green in the center of
the field of view (edges of retina not sensitive to these)
g Use black, white, yellow in peripheryg Use adjacent colors that vary in hue
& valueSep 11, 2013
IAT 814 57
Using Colorg For large regions, don’t use highly
saturated colors (pastels a good choice)
g Do not use adjacent colors that vary in amount of blue
g Don’t use high saturation, spectrally extreme colors together (causes after images)
g Use color for grouping and searchSep 11, 2013
IAT 814 58
Textureg Appears to be combination of
– orientation– scale– contrast
g Complex attribute to analyze
Sep 11, 2013
IAT 814 59
Shape, Symbolg Can you develop a set of unique
symbols that can be placed on a display and be rapidly perceived and differentiated?
g Application for maps, military, etc.g Want to look at different preattentive
aspects
Sep 11, 2013
IAT 814 60
Glyph Constructiong Suppose that we use two different visual
properties to encode two different variables in a discrete data set– color, size, shape, lightness
g Will the two different properties interact so that they are more/less difficult to untangle?– Integral - two properties are viewed holistically– Separable - Judge each dimension
independentlySep 11, 2013
IAT 814 61
Integral-Separableg Not one or other, but along an axis
Sep 11, 2013
IAT 814 62
Integral vs. Separable Dimensions
Integral
SeparableSep 11, 2013
[Ware 2000]
IAT 814 63
Shapesg Superquadric surface
can be used to display 2 dimensions
g Color could be added
Sep 11, 2013
IAT 814 64
Change Blindnessg Is the viewer able to perceive
changes between two scenes?– If so, may be distracting– Can do things to minimize noticing
changes
– http://www.psych.ubc.ca/~rensink/flicker/download/
– http://nivea.psycho.univ-paris5.fr/ECS/kayakflick.gifSep 11, 2013