Textbook Notes (270,000)
CA (160,000)
UTSC (20,000)
Psychology (10,000)
PSYB51H3 (300)
Chapter 6

PSYB51H3 Chapter Notes - Chapter 6: Stereopsis

Course Code
Matthias Niemeier

This preview shows pages 1-2. to view the full 6 pages of the document.
Chapter 6: Space Perception and Binocular Vision
Goal of visual system is to interact and perceive structure of space
Realism: world is real
Positivists: world could be fake, b/c we rely only on senses
Euclidean: geometry of real world
oi.e. parallel lines stay parallel, objects are same size when they move
oretinal images are non-Euclidean
we interpret Euclidean world from non-Euclidean images
having two eyes lets us have backup
visual field is 190 deg
binocular summation: combining signals from each eye
binocular disparity: differences between 2 retinal images
oneed for stereopsis
stereopsis: using binocular disparity as cue for depth
obetter perception of 3D
monoucular: one eye
monocular depth cues: depth cues from just 1 eye
binocular depth cue: depth cues from 2 eyes
oe.g. stereopsis
Monocular Cues to Three-Dimensional Space
impossible to perfectly recreate 3D world, so use depth cues and implicit
understanding of physics and geometry
we use multiple cues combined to understand 3D
occlusion: one object blocks view of another
oone of the most reliable depth cues, except for when we have accidental
non-metrical depth cue: only tells depth order not magnitude of depth
metrical depth cue: gives quantitative info about depth distance
Size and Position Cues
projective geometry: rules for projecting 3D to 2D
oe.g. image on retina gets smaller for farther objects
oimplicit understanding helps use depth cues
relative size: comparing size of items w/o knowing actual size
oE.g. know smaller things are farther away
Texture gradient: larger/smaller objects are grouped together and separated on

Only pages 1-2 are available for preview. Some parts have been intentionally blurred.

oE.g. see larger rabbits in one area and smaller ones in another
Relative height: using area of image to tell distance
oi.e. closer objects seen on bottom of images and farther ones at top
combine relative height and size to understand 3D
ocan see object that is same size as smaller when its at bottom of image
ogive indication of distance, but not exact
familiar size: Knowledge of typical size of objects
oe.g. big looking hand pointed towards us is probably stretched, otherwise itd
be a huge hand
ocan tell exact distance to object
relative metrical depth cues: can tell distance but only relative to other objects
oe.g. object A twice as far as B
absolute metrical depth cue: gives absolute info about distance in 3D
oe.g. nose is 4 cm in front of face
Aerial Perspective
aerial perspective (haze): knowing light is scattered by atmosphere
oe.g. farther objects more faint and bluish (shorter wavelengths scattered
Linear Perspective
linear perceptive: knowing parallel lines in 3D converge in 2D
oe.g. driving on highway looks like road is pointy (p. 142 fig. 6.15)
o lines that are parallel to plane dont converge
oRelative metrical depth info
Vanishing point: point where receding lines converge
Pictorial Depth Cues and Pictures
Pictorial depth cues: depth cues from projection of 3D to 2D
oAll cues mentioned before are pictorial
Formed by 3D world
We understand 2d picture is flat and not real, but have prediction of what it should
look like
depending on context a distorted picture will look okay or weird
oE.g. weird angle of picture doesnt look distorted if we see it on a wall. If we
just see the picture at weird angle, looks distorted.
Anamorphosis (anamorphic projection): 2D image using linear perspective so
that it only looks right when viewed at specific angle or with mirror
Motion Cues
You're Reading a Preview

Unlock to view full version