CPSC 533C Static and Moving Patterns

1
CPSC 533CStatic and Moving Patterns

• Presented by Ken Deeter
• Slides borrowed from Colin Wares PPT Slides

2
Gestalt Laws

• Gestalt German for Pattern
• Laws of pattern perception
• Proposed mechanisms were wrong
• Rules themselves still have value

3
Proximity

• Things that are close together are perceptually
  grouped together
• Rule Best way of emphasizing relationships
  between different data entities is to place them
  in proximity in a display.

4
Proximity example
5
Similarity

• Similar elements (shape, color) are grouped
  together.
• Combine with separable dimensions to assist with
  visual segmentation.
• Rule Related visual elements should look similar

6
Similarity Example
7
Continuity

• People are more likely to construct visual
  groupings out of elements that are smooth and
  continuous, rather than ones that contain abrupt
  changes in direction.
• Implies connectedness, which is stronger than
  proximity. (pg 207 fig 6.8)
• Rule use connections to show relations

8
Continuity Example
9
Connectedness Example
10
Symmetry

• Symmetrical elements emphasizes a relationship.
• Possible use use symmetry in an interface to
  emphasize similarity (figure 6.11 pg 209)
• Rule Use symmetry to relate visual elements.

11
Symmetry Example
12
Closure

• Humans tend to look for closed contours.
• Closed contours segment visual space. Organizing
  using these segmented regions is stronger than
  proximity.
• Common uses are Venn Diagrams or Application
  Windows
• Rule Use closed regions to segment visual space,
  group smaller elements.

13
Closure Examples
14
Relative Size

• Smaller components of a pattern tend to be
  perceived as objects.
• See Figure 6.15 pg 212

15
Figure and Ground

• What is foreground and what is background?
• Combination of other laws to segment image into
  fg/bg

16
Contours

• People see contours where they dont really
  exist.
• Illusory contour (figure 6.18 pg 215)

17
Transparency

• Transparency is perceived only when good
  continuity and color correspondence exists.

18
Overlap

• Perception of overlapping textures depends on
  visual interference between two patterns

19
Perceptual Syntax

• Various types of illustrations already contain
  syntax that uses Gestalt principles to code
  information.
• Node-link graphs (fig 6.29 pg 226)
• Geographical Maps (fig 6.30 pg 228)
• TreeMap (fig 6.31 pg 229)

20
Patterns in Motion

• Correspondence problem for motion to work,
  humans need to perceive correspondence of visual
  elements between different frames.
• Limit motion of elements
• Differentiate elements using shape/color/orientati
  on
• Done incorrectly, results in wagon wheel effect

21

• Vary attributes to make correlation easier.

22
Form and Contour in Motion

• Visual elements can be classified based on
  differences in movement, with comparable
  precision to static attributes.

23
Moving Frames

• Motion perception highly depend on context.
• Examples
• Synched moving dots (fig 6.34 pg 234)
• Dot in Moving frame often perceived as dot
  moving instead of frame moving.

24
Causality

• Perception of causality between two movements
  highly dependent on time between movements.
• Some experimental results suggest 160ms as
  maximum time for perceiving causality between two
  motions
• Can use specific motions to encode causal
  relationships in data.

25
Causality cont.
26
Animate Motion

• Humans are very good at perceiving
• Biological motion
• Expressiveness in motion