Depth Perception and Visualization

1
Depth Perception and Visualization

• Matt Williams

From http//www.cs.washington.edu/homes/cassidy/t
ele/index.html
2
Depth Perception and Visualization

• References and borrowed images
• Ware, C., Chapter 8 of Information Visualization
  Perception for Design. 2000, San Fancisco Morgan
  Kaufmann.
• J.D. Pfautz, Depth Perception in Computer
  Graphics, Doctoral Dissertation, University of
  Cambridge, UK, 2000.
• C. Ware, C. Gobrecht, and M.A. Paton, "Dynamic
  Adjustment of Stereo Display Parameters," IEEE
  Transactions on Systems, Man and
  Cybernetics---Part A Systems and Humans, Vol.
  28, No. 1, Jan. 1998, pp. 56-65.
• www.wlu.ca/wwwpsych/tsang/8Depth.ppt(no author
  provided)
• Robertson,G.,Mackinlay,J.,Card,S.ConeTrees
  Animated 3D visualizations of hierarchical
  information. In Proceedings of CHI'91 (New
  Orleans, LA), ACM, 189-194.
• WANGER, L., FERWANDA, J., AND GREENBERG, D. 1992.
  Perceiving spatial relationships in computer
  generated images. IEEE Computer Graphics and
  Applications (May) 44-58.

3
Depth Perception and Visualization

• Depth Perception
• Cues
• How do we combine these cues to perceive depth
• InfoVis Application
• Which cues are helpful?
• Which cues may be important in your project?

4
Depth Cues

• Monocular
• Perspective Cues
• Size
• Occlusion
• Depth of Focus
• Cast Shadows
• Shape from Motion
• Binocular
• Eye Convergence
• Stereoscopic depth

5
Structure from Motion

• Motion Parallax
• Kinetic Depth

n

• Ware, C., Chapter 8 of Information Visualization
  Perception for Design. 2000, San Fancisco Morgan
  Kaufmann.

6
Structure from Motion

• Kinetic Depth Effect
• Assumption of rigidity allows us to assume shape
  as objects move/rotate

• Ware, C., Chapter 8 of Information Visualization
  Perception for Design. 2000, San Fancisco Morgan
  Kaufmann.

7
Perspective Cues

• Parallel lines converge
• Distant objects appear smaller
• Textured Elements become smaller with distance

• Ware, C., Chapter 8 of Information Visualization
  Perception for Design. 2000, San Fancisco Morgan
  Kaufmann.

8
Perspective Cues
http//www.wlu.ca/wwwpsych/tsang/8Depth.ppt
9
Perspective Cues

• Taking advantage of linear perspective in
  visualization

• Ware, C., Chapter 8 of Information Visualization
  Perception for Design. 2000, San Fancisco Morgan
  Kaufmann.

10
Perspective Cues

• Size Constancy
• Perception of actual size versus retinal size.
• Can perceive 2D picture plane size for sketchy
  images (see below)

http//www.wlu.ca/wwwpsych/tsang/8Depth.ppt
11
Perspective Cues
http//www.wlu.ca/wwwpsych/tsang/8Depth.ppt
12
Perspective Cues
http//www.wlu.ca/wwwpsych/tsang/8Depth.ppt
13
Perspective Cues

• Usually we percieve images on the computer from
  the wrong viewpoint
• Robustness of linear perspective (Kubovy, 1986)
• e.g Movie Theatre

• Why might we want to correct for viewpoint
  changes (head movement) anyway?

• Motion Parallax
• Placement of virtual hand or object

14
Perspective Cues

• Placement of virtual hand or object
• Need for head coupled perspective

vrlab.postech.ac.kr/vr/gallery/edu/vr/display.ppt
15
Occlusion

• The strongest depth cue.

http//www.wlu.ca/wwwpsych/tsang/8Depth.ppt

• Ware, C., Chapter 8 of Information Visualization
  Perception for Design. 2000, San Fancisco Morgan
  Kaufmann.

16
Depth of Focus

• Strong Depth Cue
• Must be coupled with user input (e.g. point of
  fixation)
• Computationally expensive

• Ware, C., Chapter 8 of Information Visualization
  Perception for Design. 2000, San Fancisco Morgan
  Kaufmann.

17
Cast Shadows

• Important cue for height of an object above a
  plane
• An indirect depth cue
• Shown to be stronger than size perspective
  (Kersten, 1996)

• Ware, C., Chapter 8 of Information Visualization
  Perception for Design. 2000, San Fancisco Morgan
  Kaufmann.

18
Shape From Shading

• Ware, C., Chapter 8 of Information Visualization
  Perception for Design. 2000, San Fancisco Morgan
  Kaufmann.

http//www.wlu.ca/wwwpsych/tsang/8Depth.ppt

• Ware Chapter 7

19
Atmospheric Depth

• Reduction in contrast of distant objects
• Exaggerated in 3D displays using what is called
  proximity luminance covariance.

• Ware, C., Chapter 8 of Information Visualization
  Perception for Design. 2000, San Fancisco Morgan
  Kaufmann.

20
Eye Convergence

• Better for relative depth than for absolute depth

• Ware, C., Chapter 8 of Information Visualization
  Perception for Design. 2000, San Fancisco Morgan
  Kaufmann.

21
Stereoscopic Depth

• How it works
• Two different views fuse to one perceived view
  (try it)

• Ware, C., Chapter 8 of Information Visualization
  Perception for Design. 2000, San Fancisco Morgan
  Kaufmann.

22
Stereoscopic Depth

• Panums fusional area
• Range before diplopia occurs(worst case)
• Fovea 1/10 of a degree (3 pixels)
• Periphery 1/3 of a degree (10 pixels)
• Factors for Fusion
• Moving images
• Blurred images
• Size
• Exposure

23
Stereoscopic Depth
velab.cau.ac.kr/lecture/Stereo.ppt
24
Stereoscopic Depth
velab.cau.ac.kr/lecture/Stereo.ppt
25
Stereoscopic Depth

• Horopter
• The arc upon which everything is in the same
  location on both retinal images

26
Stereoscopic Depth

• Problems with stereoscopic displays
• Diplopia occurs when images dont fuse (try it)
• Diplopia reduced for blurred images great for
  the real world but
• Stereoscopic displays only contain sharp images.
  Close-up unattended items can be obtrusive.
• Vergence Focus Problem
• Everything on the computer screen is on the same
  focal plane.
• Causes eyestrain
• Frame Cancellation

27
Stereoscopic Depth

• Frame Cancellation

• Ware, C., Chapter 8 of Information Visualization
  Perception for Design. 2000, San Fancisco Morgan
  Kaufmann.

• Solution?

28
Stereoscopic Displays

• Cyclopean Scale
• Move virtual environment close to the display
  plane
• No Cancellation
• Reduced Vergence-focus problem

• Ware, C., Chapter 8 of Information Visualization
  Perception for Design. 2000, San Fancisco Morgan
  Kaufmann.

29
Stereoscopic Displays

• Virtual Eye Separation (Telestereoscope)
• Allows for a decrease or increase in disparity
• Allows for an increase or decrease in the depth
  of the virtual environment

http//www.cs.washington.edu/homes/cassidy/tele/in
dex.html
30
Depth Perception Theory

• General Unified Theory
• Perceived Depth Weighted sum of all Depth Cues
• Rank the cues in importance
• e.g.
• Occlusion
• Motion Parallax
• Stereo
• Size constancy
• Etc.

31
Depth Perception Theory

• Importance changes with distance

, 96
32
Space Perception Theory

• Task Dependant Model
• Cues weights are combined differently based on
  the task
• Evidence?
• Task Orientation of a virtual Object
• Cast Shadows and Motion Parallax help
• But Linear Perspective hinders such orientation
• Task Object translation
• Linear perspective was the most useful cue

Wanger, 1992
33
InfoVis Tasks

• Tracing 3D data paths
• Judging 3D surfaces
• Finding 3D patterns of points
• Relative Position in 3D space
• Judging movement of Self
• Judging Up Direction
• Feeling a sense of Presence

34
Tracing 3D Data Paths

• Benefits of 3D Trees
• More nodes can be displayed (Robertson et al.,
  1993)
• Reduced errors in detecting Paths (Sollenberger
  and Milgram, 1993)

• Ware, C., Chapter 8 of Information Visualization
  Perception for Design. 2000, San Fancisco Morgan
  Kaufmann.

35
Tracing 3D Data Paths

• Beneficial Cues
• Kinetic Depth and Stereoscopic Depth reduced
  errors in path detection
• Kinetic Depth was the stronger cue
• Occlusion Is helpful
• (Ware and Franck, 1996)

36
Tracing 3D Data Paths

• Cone Trees
• (Robertson, 1993)

37
Judging 3D surface structure

• Judging height, gradient, curvature etc.
• Judging height of Cones
• Stereo depth gt Structure from motion (Durgen et
  al, 1995)
• Judging the gradient of textured surface
• Structure from motion gt Stereo (Tittle et al.,
  1995)
• The importance of Stereo Depth, Kinetic Depth,
  Shape from Shading, Surface Textures is situation
  dependant.
• Conclusion Arbitrary surfaces - Include them all
  

38
3D Patterns of Points
http//www-pat.fnal.gov/nirvana/plot_wid.html
http//neutrino.kek.jp/kohama/sarupaw/sarupaw_htm
l/fig/nt_3d.gif
39
3D Patterns of Points

• Beneficial Cues
• Structure from motion
• Stereo Depth
• Not Beneficial
• Perspective
• Size
• Cast Shadows
• Shape from Shading (How?)

40
3D Patterns of Points

• Add shape to clouds of points

• Ware, C., Chapter 8 of Information Visualization
  Perception for Design. 2000, San Fancisco Morgan
  Kaufmann.

41
Judging Relative Position

• Small Scale (Threading a needle)
• Beneficial Stereo
• Not Beneficial Motion Parallax
• Large Scale ( gt 30 m)
• Beneficial motion parallax, perspective, cast
  shadows, texture gradients
• Not Beneficial stereo

• Ware, C., Chapter 8 of Information Visualization
  Perception for Design. 2000, San Fancisco Morgan
  Kaufmann.

42
Movement of Self (Vection)

• Parameters
• Field Size
• Background motion (Train example)
• Stereo will help determine if something is
  perceived as background

43
Feeling of Presence

• Beneficial Parameters
• High frame rate
• High level of detail
• Not Beneficial
• Stereo (did not add to realism) (Hendrix and
  Barfield, 1996)

44
Conclusion

• Depth Cues
• Existing Theories
• Application to InfoVis

• Occlusion
• Texture Gradient
• Size Constancy
• Cast Shadows
• Stereo

From http//www.cs.washington.edu/homes/cassidy/t
ele/index.html
45
Discussion

• Projects??
• Monocular
• Perspective Cues
• Size
• Occlusion
• Depth of Focus
• Cast Shadows
• Shape from Motion
• Binocular
• Eye Convergence
• Stereoscopic depth