2018-02-05 Depth Perception PSY 525.001 Vision Science 2018 Spring - - PowerPoint PPT Presentation

▶
2018 02 05 depth perception
SMART_READER_LITE
LIVE PREVIEW

2018-02-05 Depth Perception PSY 525.001 Vision Science 2018 Spring - - PowerPoint PPT Presentation

Jul 02, 2023 153 likes •965 views

2018-02-05 Depth Perception PSY 525.001 Vision Science 2018 Spring Rick Gilmore 2018-02-05 08:53:13 1 / 77 2 / 77 But what is the Fourier Transform? A visual introduction. 3 / 77 Today's topics 4 / 77 Today's topics Discuss

slide-1
SLIDE 1

2018-02-05 Depth Perception

PSY 525.001 • Vision Science • 2018 Spring

Rick Gilmore 2018-02-05 08:53:13

1 / 77

slide-2
SLIDE 2

2 / 77

slide-3
SLIDE 3

But what is the Fourier Transform? A visual introduction.

3 / 77

slide-4
SLIDE 4

Today's topics

4 / 77

slide-5
SLIDE 5

Today's topics

Discuss project proposal

4 / 77

slide-6
SLIDE 6

Today's topics

Discuss project proposal Depth perception

4 / 77

slide-7
SLIDE 7

Today's topics

Discuss project proposal Depth perception Discuss Leopold & Logothetis, N. K. (1996).

4 / 77

slide-8
SLIDE 8

Term project

5 / 77

slide-9
SLIDE 9

http://psu-psychology.github.io/psy-525-vision-spring-2018/project- proposal.html 6 / 77

slide-10
SLIDE 10

Perceiving surfaces orientation (spatial layout)

Slant, tilt, distance

7 / 77

slide-11
SLIDE 11

8 / 77

slide-12
SLIDE 12

Marr's 2.5 D sketch

9 / 77

slide-13
SLIDE 13

10 / 77

slide-14
SLIDE 14

Depth perception

11 / 77

slide-15
SLIDE 15

"For those of a creationist bent, one could note that God must have loved depth cues, for He made so many of them" Yonas & Granrud, 1985, p. 45 12 / 77

slide-16
SLIDE 16

Kinetic Binocular Static (pictoral) Sensorimotor

13 / 77

slide-17
SLIDE 17

Monocular (kinetic, static, some sensorimotor) vs. binocular cues

14 / 77

slide-18
SLIDE 18

Kinetic cues to depth

15 / 77

slide-19
SLIDE 19

Kinetic Depth Effect Demo

Kinetic depth eect (KDE)

16 / 77

slide-20
SLIDE 20

By Nobuyuki Kayahara - Procreo Flash Design Laboratory, CC BY-SA 3.0, Link This one is also bistable 17 / 77

slide-21
SLIDE 21

Kinetic depth eect

3D structure perceived from temporal sequence of 2D (outline-only) views. Reported in Wallach, H., & O’Connell, D. N. (1953). The Kinetic Depth Effect. Journal of Experimental Psychology, 45(4), 205. 18 / 77

slide-22
SLIDE 22

Structure from Motion Demo

Structure from motion

19 / 77

slide-23
SLIDE 23

Archival Gibson - 1958 - Motion parallax and perceived depth

Motion parallax

20 / 77

slide-24
SLIDE 24

City motion parallax

21 / 77

slide-25
SLIDE 25

The geometry of motion parallax

Where is fixation? What is the direction of motion? 22 / 77

slide-26
SLIDE 26

PJ Treffner - Occlusion demos: Accretion and deletion

Texture accretion & deletion

23 / 77

slide-27
SLIDE 27

Second-order motion perception

Texture accretion & deletion

24 / 77

slide-28
SLIDE 28

Task 1: Example of 100% Radial Optic Flow (no random dots) wit…

Optic ow

25 / 77

slide-29
SLIDE 29

AR.Drone Helicopter Optical Flow Test #1

Optic ow and autonomous ight

26 / 77

slide-30
SLIDE 30

Binocular cues to depth

27 / 77

slide-31
SLIDE 31

Stereopsis

Perception of depth and 3D structure from stimulation of both eyes 28 / 77

slide-32
SLIDE 32

Horopter

a line or surface containing all those points in space whose images fall on corresponding points of the retinas of the two eyes. 29 / 77

slide-33
SLIDE 33

30 / 77

slide-34
SLIDE 34

Random-dot stereograms

Invented by Béla Julesz 31 / 77

slide-35
SLIDE 35

32 / 77

slide-36
SLIDE 36

library(imager) n_pts <- 100 left_img <- array(round(runif(n=n_pts^2),0), dim = c(n_pts, n_pts)) plot(as.cimg(left_img), axes=FALSE)

33 / 77

slide-37
SLIDE 37

right_img <- left_img square_pix <- 20 center_square <- right_img[floor(n_pts/2-square_pix/2):floor(n_pts/2+ plot(as.cimg(center_square), axes=FALSE)

34 / 77

slide-38
SLIDE 38

Auto-stereogram

Can't really fuse these from projected image. Why? 35 / 77

slide-39
SLIDE 39

The "correspondence" problem

How do retinal image points/edges correspond to object points/edges? Why can it take time to "fuse" stereograms? 36 / 77

slide-40
SLIDE 40

37 / 77

slide-41
SLIDE 41

Marr-Poggio algorithm for solving

Use: surface opacity & surface continuity heuristics Iterate until a best-fitting solution is found 38 / 77

slide-42
SLIDE 42

Da Vinci stereopsis

Different eyes see different portions of surfaces 39 / 77

slide-43
SLIDE 43

Geometry of self-motion, object-motion, and disparity

40 / 77

slide-44
SLIDE 44

Static (pictoral) cues to depth

41 / 77

slide-45
SLIDE 45

Linear Perspective

42 / 77

slide-46
SLIDE 46

43 / 77

slide-47
SLIDE 47

44 / 77

slide-48
SLIDE 48

Public Domain, Link

Linear perspective + elevation over horizon = Ponzo illusion

45 / 77

slide-49
SLIDE 49

Relative size

46 / 77

slide-50
SLIDE 50

47 / 77

slide-51
SLIDE 51

48 / 77

slide-52
SLIDE 52

Ames Room Illusion

Ames room

49 / 77

slide-53
SLIDE 53

Actual position of Person A Apparent position

  • f person A

Apparent shape of room Viewing peephole Actual and apparent position

  • f person B

https://en.wikipedia.org/wiki/Ames_room 50 / 77

slide-54
SLIDE 54

Aerial perspective Defocus blur

51 / 77

slide-55
SLIDE 55

By Joaquim Alves Gaspar - Own work, CC BY-SA 2.5, Link 52 / 77

slide-56
SLIDE 56

Interposition, occultation

53 / 77

slide-57
SLIDE 57

Texture gradients

54 / 77

slide-58
SLIDE 58

By Gustave Caillebotte - 5wEUCOlEf-EaVQ at Google Cultural Institute maximum zoom level, Public Domain, Link 55 / 77

slide-59
SLIDE 59

56 / 77

slide-60
SLIDE 60

http://psychsciencenotes.blogspot.com/2011/08/mirrors-are-literally-windows- to.html 57 / 77

slide-61
SLIDE 61

Lighting, shading, & shadow cues

58 / 77

slide-62
SLIDE 62

59 / 77

slide-63
SLIDE 63

60 / 77

slide-64
SLIDE 64

61 / 77

slide-65
SLIDE 65

Sensorimotor cues to depth

Palmer's "ocular" cues

62 / 77

slide-66
SLIDE 66

Vergence

Eyes (typically) converge on a 3D point. Angle of vergence related to 3D geometry. 63 / 77

slide-67
SLIDE 67

Lens Accommodation

Accommodation

64 / 77

slide-68
SLIDE 68

<

The accommodation reex

Vergence, pupil diameter change, + accommodation

65 / 77

slide-69
SLIDE 69

2-Minute Neuroscience: Vestibular System

Vestibular system detects head rotation, translation

66 / 77

slide-70
SLIDE 70

Vestibular signals speed, direction of rotation, translation More motion parallax with head translation than rotation

67 / 77

slide-71
SLIDE 71

Comparing the cues

68 / 77

slide-72
SLIDE 72

The human brain in depth: how we see in 3D

Integrating the cues

Welchman, A. E. (2016). The human brain in depth: how we see in 3D. Annual Review of Vision Science. annualreviews.org. Retrieved from http://www.annualreviews.org/doi/abs/10.1146/annurev-vision-111815-114605 69 / 77

slide-73
SLIDE 73

Heuristics

About world, illumination conditions

70 / 77

slide-74
SLIDE 74

Alternative view:

These aren't cues; they are information. Animals don't "reconstruct" 3D space; they perceive it directly. 71 / 77

slide-75
SLIDE 75

Amazing T-Rex Illusion!

72 / 77

slide-76
SLIDE 76

Break time

73 / 77

slide-77
SLIDE 77

Leopold & Logothetis, N. K. (1996)

74 / 77

slide-78
SLIDE 78

Core phenomena

Binocular rivalry Neural basis of binocular rivalry Neural basis of "conscious" visual experience 75 / 77

slide-79
SLIDE 79

Next time...

Perceptual organization Size, shape, orientation, & position

76 / 77

slide-80
SLIDE 80

Slides created via the R package xaringan. Rendered HTML and supporting files are pushed to GitHub where GitHub's 'pages' feature is used to host and serve the course website. 77 / 77