Depth Perception LOs Flashcards
How does visual system obtain depth information
Physiological depth cues for distances less than about 3m
Accomodation: crystalline lens gets fatter to focus on nearby objects and gets thinner to focus on farway objects, feedback from ciliary muscles give info on lens curvature to get distance
Convergence: rotation of eyes inward to cause image to fall on the fovea, large convergence angle = closer object, smaller convergence angle - further
binocular/retinal disparity: retinal images of an object fall on disparate points on each eye’s retina, fixated object produces no disparity
Horopter: imaginary surface passing through fixation point; retinal images of objects not on horopter fall on disparate points on each eyes retina
Distance between points on each retina is the degrees of disparity, farther from horopter the greater the disparity
Panum’s fusional area (PFA): spatial points that fall on non corresponding retinal areas, but lie within the PFA are fused into single images
How is retinal disparity related to depth
The greater the degree of disparity the farther from the horopter
stereopsis: perception of depth based on retinal disparity alone. disparity = difference in images from two eyes
in things both on horopter than approximately same distance away
can have crossed and uncrossed disparity (crosses equals in front of horopter and closer to observer uncrossed is opposite)
What is the correspondence problem
How is it determined which elements in the two images match and fuse together
2 solutions: feature-based methods make a match based on extracted image structure, correlation-based methods are based on grey-level descriptions (statistically likely to match)
Spring-coupled magnetic dipole model makes global match: to induce fusion both images must lie on horpter so only neurons responding to 0 degrees disparity are involved, after images have fused you can move them off horopter without losing fusion within PFA (hysteresis effect) - ONLY WORKS WITH RANDOM DOT STEREOGRAMS NOT NATURAL IMAGES
What natural constraints for depth are exploited by Marr and Poggio’s programs
Their program elaborated on idea of cooperating neurons
ASSUMES: both eyes looking at same object, any point can occupy only one place at any time, matter is cohesive and objects tend to be uniform
Compatibility: to be matched points on each retina must be physically similair people unable to fuse an image with same image of reversed contrast (eg. dark features in one correspond to dark features in the other)
Uniqueness (opacity): a feature on one retina should correspond uniquely to one feature on the other retina, violated in the case of transparent surfaces, when an image feature is a combo of points from two physical surfaces (ie. looking through a glass violates this)
Continuity: disparity should vary smoothly, if two matched features are close together in images then typically their disparities will be similar because environment is made of continuous surfaces separated by boundaries, changes in disparity should be rare (Only at surface boundaries)
Marrs model is 2 stages: 1) all possible matches are established and represented in a neural network of connections (via compatibility constraint). 2) false matches are inactivated by competition from true targets: uniqueness produces mutual inhibition of matches along a line of sight, continuity constraints produces mutual excitation of matches having same/similar depth
Describe Marr’s 2 ½ D Sketch
It is used to help obtain stereopsis despite blurring
Not a full 3D model of the world, represents surface orientation
Based on: full primal sketch (contour, texture, shading, occlusion depth cues), stereopsis (retinal disparity), shape from motion (rotation of object)
Represented by vectors: give orientation of surface at a given point, represent amount of slant
If not enough info to create 2 ½ D sketch, certain natural constraints must be applied
Characteristics: contains contour, texture and shading info, incorporates depth via linear perspective/retinal disparity, basic figure/ground segregation begun (relative depth not absolute), representation is viewed centered (depends on your location as the viewer)
How are visual illusions different from adaptation effects
Adaptation effects are successive illusions, after effects that require an adapting phase
Visual illusions are simultaneous: interacting stimuli cause the effect
What can illusions tell us about functional architecture
Illusions can test where the underlying process are apart of the FA using the cognitive penetrability criterion
Methods: 1) test perceptual phenomenon 2) change a belief related to the effect (top-down) 3) retest
If effect has changed it is cognitively penetrable and the underlying processes are NOT part of the FA
What are some natural constraints that lead to the perception of some illusions
Misapplied size constancy: as your distance from a given object increases, the retinal image becomes smaller, yet we do not perceive the object as getting smaller (size constancy). Distance is taken into account when perceiving size (size-distance scaling)
eg. Ponzo illusion: converging lines are a depth cue that activates size constancy; 2-D image treated like a 3-D image. Visual angles subtended by both horizontal bars are identical, because of size-distance scaling the upper bar seems farther away and thus seems larger
Visual experience:
Muller-Lyer Illusion: two lines with inward arrows or outward arrows, weaker in children and people living in dome-shaped huts, possible carpentered world hypothesis (like inside corner of a room (outward arrows) or like outside edge of a building (inward arrows). Depth cues indicate outward arrows relatively far away and inward are relatively close so due to size constancy outward appears longer and inward appears shorter
Constancy and Visual Experience Illusions:
Mcgraw (2003): size constancy explanation of Poggendorff illusion.
-Looks like the line doesn’t add up but it actually does!
Consider it to consist of two separate components (left and right side) - each side if similar to the Muller-Lyer Illusion
Upper segment appears shorter than it is and lower segment appears longer than it is
Left segment appears lower than it is and right segment appears higher than it is
Cortical area:
Ebbinghaus illusion: circle surrounded by big circles looks smaller compared to circle surrounded by small circles: the large the size of V1 in an individual can cause a weaker illusion
