Space Perception and Binocular Vision Flashcards
Positivism
Phitosophical position arguing that all we really have to go off of is the evidence of our senses, so world might be nothing more than an elaborate hallucination.
Realism
Argues theres a real world to sense.
Looking out at the world
2D with a sense of 3D.
Fine taking in parts, but confused when we look at it as a whole.
Put together contours to form images.
How do we know how far something is?
We use perception to interact with world.
We need to know where things are relative to us, to determine how we will interact with them.
3D>2D>3D
Problem of the visual system needs to construct a 3D world based on inverted images on retina of each eye.
Goal of depth perception: to accurately perceive a 3D
world on the basis of two 2D retinal images (one in each
eye).
our retina is not flat( is a curved surface).
2 Retinal images always differ because eyeballs are in a different position on head.
Retinal area occupied by an object gets smaller as an object moves away from eyeball. -If we want to understand a 3D world we have to reconstruct of distorted retinal input.
Angles are often greater.
Challenge: an infinite
number of different 3D
scenes can produce the
same retinal image.
Probability Summation
Increased detection probability based on statistical advantage of having two (or more) detectors than one
Binocular Summation
Probability summation in vision.
Increased detection because of two eyes
Pictorial depth cue
A cue to distance or depth used by artists to depict 3D depth in 2D pictures
Trick to depict depth in their painting
Monocular depth cue
a depth cue that is available
even when the world is viewed with one eye alone
patch over or damaged
Binocular depth cue
a depth cue that relies on
information from both eyes
pick up more information from environment
Eyes let us see more of the world– Especially true for animals that have lateral ones. ie/ rabbit. Good for animals of prey
Humans have frontal eyes
Ie/ stereopsis
Oculomotor Depth Cues
Cues that are based on
feedback from the oculomotor muscles controlling the shape of the lens and the position of the eyes
The way the eyes move to get a clear image.
Where the eye are actually moving to get info on fovea, Muscle focused;
Accommodation (lens).
Convergence/Divergence.
Accommodation
the process by which the eye changes its focus (lens changes its shape).
Contraction of ciliary muscles for near objects (lens gets fatter)–lens more round, bending to get focus on back of eye.
Relaxation of ciliary muscles for far objects.-flatter lens.
Only provides depth information for objects <2m away
Lens has a limit for how much it can change shape.
Convergence
the ability of the two eyes to turn inward, often used to focus on nearer objects
Binocular cue–so need to compare. with one eye could just be looking to the side.
are eyes turning inward or outward.
Divergence
The ability of the two eyes to turn outward, often used to focus on farther objects
Binocular cue–so need to compare. with one eye could just be looking to the side.
are eyes turning inward or outward.
static monocular depth cue
Static monocular depth cues: cues that provide
information abut depth on the basis of:
Position of objects in the retinal image,
Occlusion.
Relative height.
Size of objects in the retinal image.
Relative size.
Familiar size.
The effects of lighting in the retinal image.
Shadows/Shading.
Aerial (atmospheric) perspective.
Non metrical depth cue
a depth cue that provides info about the depth cue order (relative depth) but not depth magnitude (nose is in front of his face).
misleading only in case of accidental viewpoints.
Metrical depth cue
a depth cue that provides quantitative info about distance in 3rd dimension.
Familiar size
a cue based on knowledge of the typical size of objects.-what size ought to be
often works in conjunction with the cue of relative size.
Relative metrical depth cue
Could specify that object A is twice as far away as object B without providing info about the absolute distance to either A or B
relative sight and high do not tells us the exact distance to/between objects.
Absolute metrical depth cue
Provides quantifiable info about distance in the 3D
familiar size
Projective geometry
Geometry that describes the transformations that occur when 3D world is projected onto a 2D surface.
Ie/ parallel lines do not converge in the real world, but they do in 2D projection of that world.
Aerial (or atmospheric) perspective. or haze
a depth cue based on the implicit understanding that light is scattered by the atmosphere.
Short wavelengths (blue) are scattered more than medium and long wavelengths. Why the sky looks blue.
More light is scattered when we look through more atmosphere
thus, more distant objects appear fainter, bluer, and less distinct.
contours become less distinct as we get further away.
(Monocular depth cues)
Linear perspective
Lines that are parallel in the three-dimensional world will appear to converge in a two-dimensional image as they extend into the distance
(Monocular depth cues)
Vanishing point
The apparent point at which parallel lines receding in depth converge.
(Monocular depth cues)
Relative height
(“height in plane”, “proximity to horizon”):
Objects touching the ground: higher = farther
Objects in the sky (above horizon): lower = farther
Objects will be higher in the visual field if they are more distant.
Higher it is the further away from ground plane.
but closer clouds in sky plane.
Closer=Higher
Size-distance relation
Size-distance relation: the farther away an object is from the observer, the smaller is its retinal image
Relative size
A comparison of size between items without knowing the absolute size of either one.
all things being equal, we assume that smaller objects are further away from us than larger objects.
Texture gradient
a depth cue based on the geometric fact that items of the same size form smaller, closer spaced images the farther away they get.
Result from a combination of the cues of relative size and relative height.
as you go up in image.
Space gets closer together as you get further away.
(Monocular depth cues)
▪ Anamorphosis (or anamorphic projection)
use of the rules of linear perspective to create a two-dimensional image so distorted that it looks correct only when viewed from a special angle or with a mirror that counters the distortion.
our ability to cope with distortion is limited.
rules of linear perspective are pushed to an extreme.
projection of 3D to 2D.
(Monocular depth cues)
Dynamic monocular depth cues (Monocular depth cues)
cues that provide information abut depth on the basis of motion.
Motion parallax.
optic flow.
Deletion and Accretion.
(Monocular depth cues)
triangulation
triangle formed by the two eyes and the point they fixate on in the 3D world.
Motion parallax
images closer to the observer move faster across the visual field than images farther away.
Head movements and any other relative movements between observers and objects reveal motion parallax cues.
A triangulation cue
Based on head movement– Allow for seeing multiple viewpoints by moving head.
Animals use it for prey to figure out when to attack.
Monocular depth cues
Optic flow
the changing angular position of points in a perspective image that we experience as we move through the world.
The pattern of apparent motion of objects in a visual scene produced by the relative motion between the observer and the scene.
Monocular depth cues
Deletion
the gradual hiding (occlusion) of an object as it passes behind another one.
Object will be further away from you in order far this to happen.
Monocular depth cues
Accretion
Object will be further away from you in order far this to happen.
the gradual revealing (“de-occlusion”) of an object as it emerges from behind another one.
Monocular depth cues
Binocular Vision
The probability of detecting a stimulus increases with more samples.
One of the advantages of having two eyes that face forward.
The two retinal images of a 3D world are not the same.
Animals with smaller brains extract depth from disparity.
Key points;
▪ Horopter is always centered on fixation point.
▪ Objects on the horopter have zero binocular disparity.
▪ Binocular disparity increases as you move away from the horopter.
▪ If closer than the horopter disparity is crossed.
▪ More crossed disparity as the object gets closer.
▪ If further than the horopter disparity is uncrossed.
▪ More uncrossed disparity as the object gets further.
In order to perceive depth from brain must know which part of the right retinal image corresponds with which part of the left retinal image
input from two must converge onto the same neuron - this convergence doesn’t happen until the V1
V1 neurons are binocular; can be excited by either eye
V2
Computes depth order based on contour completion and border ownership.
Intermediate visual areas (Ie/V4)
Encode depth intervals, based on relative disparities.
Higher cortic areas (inferotemporal cortex)
Are involved in representation of complex 3D shape
Binocular neurons in VI
are most responsive to absolute disparities
Input from two us must converge onto the same neuron–This convergence doesn’t happen until the vI
vI neurons are binocular; can be excited by either eye
Neurons in V2 and many other higher cortical areas
are sensitive to relative disparities
Input from two us must converge onto the same neuron–This convergence doesn’t happen until the vI
vI neurons are binocular; can be excited by either eye
Relative diparties
Provide the basis for very fine stereoaculty.
VI signals feed into both ventral(more complex) and dorsal streams(control eye movements)
The difference in the absolute disparities of two objects.
binocular disparity
the differences between the two retinal images of the same scene.
Disparity is the basis for stereopsis, a vivid perception of the three-dimensionality of the world that is not available with monocular vision. Objects “popping out”
Effective up to 200 m
Observers are very sensitive to small binocular disparities
Input from two us must converge onto the same neuron–This convergence doesn’t happen until the vI
vI neurons are binocular; can be excited by either eye
Middle temporal area
Important in perception of motion.
Cells here can signal sign of depth based on motion parallax of shaking your head.
Stereoaculty
is a measure of the smallest binocular disparity that can generate a sense of depth.
Increased after stereopsis rapidly until adult levels
rose from nothing first 4 months to near adult levels by 6 months
different from simple acuity that takes years to develop.
absolute disparity
the difference in the angular distance of the images of an object from the foveas of the two eyes.
Not consciously available ti make perceptual judgments
disregarded because convergence noise is common whereas relative disparities are not.
Corresponding points
a point on the left retina and a point on the right retina that would coincide if the two retinas were superimposed (e.g., the foveas of the two eyes)
if they are at the same distance to the fovea on both sides.
tend to have zero binocular disparity.
binocular vision
Non-corresponding points
a point on the left retina and a point on the right retina that would not coincide if the two retinas were superimposed (e.g., the fovea of one eye and a point 4 mm to the right of the fovea in the other eye)
objects significantly closer to or farther away from the surface of zero disparity form images on non corresponding parts of the eyes.- this double vision is known as diplopia.
binocular vision.
Occlusion
a cue to relative depth order in which,
for example, one object partially obstructs the view
of another object.
gives info about relative position of objects.
argues to be most reliable of all depth cues.
more powerful depth cue.
light needs to come from object to your eye to see it, so if something is blocking/obstructing will reach your eye
can be misleading in the care of accidental viewpoints
ie/ weirdly fitting puzzle pieces
more likely to be the generic view
blocking=closer
Horopter
the location of objects whose images lie on the corresponding points in the two retinas.
The surface of zero disparity
Objects on the horopter are seen as single images when viewed with both eyes.
The Vieth-Müller circle and the horopter are technically different, but for our purposes you may consider them the same.
Panum fusional area
the region of space, in front of and behind the horopter, within which binocular single vision is possible.
Diplopia
Double vision
If visible in both eyes, stimuli falling outside of Panum’s fusional area will appear diplopic.
Objects significantly closer to or farther away from the horopter fall on non-corresponding points in the two eyes and are seen as two images.
will be experienced if an adult becomes estropic
Crossed disparity
the sign of disparity created by
objects in front of the plane of the horopter. (fixation)
images in front of the horopter are displaced to the
left in the right eye and to the right in the left eye
things closer
Uncrossed disparity
the sign of disparity created by
objects behind the plane of the horopter
images behind the horopter are displaced to the right
in the right eye and to the left in the left eye
created by things further away from you than horropter
when we move things to the distance
Correspondence problem
in binocular vision, the problem of figuring out which bit of the image in the left eye should be matched with which bit in the right eye
solving all the time if you have normal binocular vision
Particularly vexing in random dot stereograms
consisting of many similar features
1) (at least) two options:
Object recognition performed separately on the information from each eye, then the objects are matched and disparity is determined (object recognition precedes correspondence matching)
2) The visual system matches the retinal images based on simple features before object recognition (correspondence matching precedes object recognition)
▪ Test: Bela Julesz: random dot stereograms
Binocular rivalry
Competition between two eyes for control of visual perception.
never completely won by either eye or either stimulus.
evident when completely different stimuli are presented to the two eyes.
visual system chooses to surpress an image and perceive the other.
occurs when corresponding points in the eyes are unrelated
dichoptic
presentation of two different stimuli, one to each eye
cyclopean
referring to stimuli that are defined by binocular disparity alone.
Stereoscope
a device for presenting one image to one eye and another image to the other eye.
can be used to present dichoptic stimulus for stereopsis and binocular rivalry.
confirmed that the visual system treats binocular disparity as a depth cue.
binocular vision
Random dot stereogram (RDS)
a stereogram made of a large number of randomly placed dots.
to prove that stereopsis might be used to discover objects and surfaces in the world.
contain monocular cues to depth and no “objects”
thought it might help reveal camouflaged objects.
cats have stereopsis.
binocular vision
Free fusion
the technique of converging (crossing) or diverging (uncrossing) the eyes in order to view a stereogram without a stereoscope.
“the poor man’s stereoscope”
The phenomenon of double vision (diplopia).
critical period
a period of time in development when the organism is particularly susceptible to developmental change.
Strabismus
a misalignment of the two eyes such that a single object in space is imaged on the fovea of the one and on nonfoveal are of other (turned eye)
esotropia
strabismus in which one eye deviates inward (“cross eyed”)
exotropia
strabismus in which one eye deviates outward (“wall eye”)
Significance of random dot stereograms
If object recognition necessarily precedes correspondence matching, no depth should be seen in a random-dot stereogram, because it does not contain any objects
But, since we do see depth in a random-dot stereogram, correspondence matching must precede object recognition
Uniqueness constraint
the observation that a feature in the world is represented exactly once in each retinal image
binocular vision
Continuity constraint
the observation that, except at the edges of objects, neighboring points in the world lie at similar distances from the viewer
binocular vision
How is stereopsis implemented in the human brain?
Input from two eyes must converge onto the same cell.
Many binocular neurons respond best when the retinal images are on corresponding points on the two retinas.
neural basis for horopter.
Many other binocular neurons respond best when similar images occupy slightly different positions on the retinas of the two eyes.
turned to particular disparity.
Development of Stereopsis
onset by 3-5 months.
adult levels within 6-7 months.
abnormal visual experience can disrupt binocular vision:
Strabismus
a misalignment of the two eyes such that a single object in space is imaged on the fovea of one eye, and on the nonfoveal area of the other (turned) eye.
▪ Can lead to the suppression of the image from the mis-aligned eye -> stereoblindness
Amblyopia
a developmental disorder characterized by reduced spatial vision in an otherwise healthy eye (“lazy eye”)
tilt aftereffect
the perceptual illusion of tilt, produced by adaptation to a pattern to a pattern of given orientation.
Stereoblindness
an inability to make use of binocular disparity as a depth cue.
many people who are stereoblind do not even realize it.
Rambrandt was probably stereoblind
3-5 % of population
usually a secondary effect of childhood visual disorders
Bayesian approach
estimates probability of a current observation.
formulating the idea that out perception is a combo of current stimulus.
which interpretation seems more likely.
automanic.
Ie/ pennies are all the same size in our experience.
Combining depth cues
depth cues need to be combined
Our perception is a combination of the current stimulus and our knowledge about the conditions of the world – what is and is not likely to occur.
The Bayesian approach
Prior knowledge can influence our estimates of the probability of an event.
Size constancy
Perceived distance
distance/depth cues
generally, the more depth cues, the more accurate our size constancy
▪ In all of the previous illusions, R was the same and perceived D changed resulting in the illusion.
▪ What if perceived D is kept constant and R changes?
the perception of an object as having a constant size even though our sensation of the object changes.
moon illusion
perceive the moon to be larger when closer to the horizon.
retinal size the same.
perceived distance differs.
Emmert’s Law
S=RxD
Ponzo illusion
over interpret the depth cues in a 2D image
Ames illusion
Size of people looks distorted because you are looking at distorted room.
Perceptual constancies
Objects appear to be:
▪ The same size when viewed from different distances
▪ The same shape when viewed from different angles
▪ The same colour when viewed in different lighting
Shape constancy
Edges within an object change their relative distance to us as we rotate the object or move relative to it.
Likewise, the retinal image changes
the relative constancy of the perceived shape of an object despite variations in its orientation
