TANGENT ALERT What happens when no correspondence is

  • Slides: 20
Download presentation
TANGENT ALERT! What happens when no correspondence is possible? Highly mismatched stereo-pairs lead to

TANGENT ALERT! What happens when no correspondence is possible? Highly mismatched stereo-pairs lead to ‘binocular rivalry’ Open question: Can rivalry and fusion coexist?

Computational theories for solving the correspondence problem: Given the underconstrained matching problem (100! Possible

Computational theories for solving the correspondence problem: Given the underconstrained matching problem (100! Possible pairings in an RDS with 100 dots), what assumptions can we bring to bear? Assumption 1: Epipolar constraint

Marr-Poggio’s network-based formulation of the problem: Assumptions: 1. 2. 3. Surface opacity / match

Marr-Poggio’s network-based formulation of the problem: Assumptions: 1. 2. 3. Surface opacity / match uniqueness Surface continuity Match compatibility

Sample result of Marr-Poggio’s network:

Sample result of Marr-Poggio’s network:

Enhancing the Marr-Poggio’s model: Edge-based matching rather than pixel matching. Advantages: 1. Edge orientation

Enhancing the Marr-Poggio’s model: Edge-based matching rather than pixel matching. Advantages: 1. Edge orientation and polarity provide additional matching constraints 2. Greater consistency with known physiology (matching begins in V 1) Disadvantages:

Enhancing the Marr-Poggio’s model: Edge-based matching rather than pixel matching. Advantages: 1. Edge orientation

Enhancing the Marr-Poggio’s model: Edge-based matching rather than pixel matching. Advantages: 1. Edge orientation and polarity provide additional matching constraints 2. Greater consistency with known physiology (matching begins in V 1) Disadvantages: 1. Depth information is sparse; an additional process of interpolation is is needed.

Enhancing the Marr-Poggio’s model: Edge-based matching rather than pixel matching. Advantages: 1. Edge orientation

Enhancing the Marr-Poggio’s model: Edge-based matching rather than pixel matching. Advantages: 1. Edge orientation and polarity provide additional matching constraints 2. Greater consistency with known physiology (matching begins in V 1) Disadvantages: 1. Depth information is sparse; an additional process of interpolation is is needed. Open problems: 1. How to match stereo pairs where assumptions are violated? 2. How to make use of monocular shape cues?

Physiological mechanisms of stereopsis: Hubel and Wiesel (1962): Binocular cells in V 1 not

Physiological mechanisms of stereopsis: Hubel and Wiesel (1962): Binocular cells in V 1 not sensitive to disparity (in cats) Barlow et al (1967): V 1 cells sensitive to disparity Hubel and Wiesel (1970): V 1 cells not sensitive but V 2 cells are (monkeys) Poggio and Fischer (1977): V 1 cells sensitive to small disparities and V 2 cells sensitive to large disparities (awake fixating monkeys)

Cue integration:

Cue integration:

Processing Framework Proposed by Marr Recognition 3 D structure; motion characteristics; surface properties Shape

Processing Framework Proposed by Marr Recognition 3 D structure; motion characteristics; surface properties Shape From stereo Motion flow Shape From motion Color estimation Edge extraction Image Shape From contour Shape From shading Shape From texture

Motion Perception: -Detecting motion and motion boundaries -Extracting 2 D motion fields -Recovering 3

Motion Perception: -Detecting motion and motion boundaries -Extracting 2 D motion fields -Recovering 3 D structure from motion

Motion as space-time orientation:

Motion as space-time orientation:

Computational models of motion detectors: Delay and compare networks

Computational models of motion detectors: Delay and compare networks

Other ways of constructing movement detectors: Psychophysical support from Anstis’ experiment (1990)

Other ways of constructing movement detectors: Psychophysical support from Anstis’ experiment (1990)

TANGENT ALERT! Accounting for eye-motion Q. When do we see an object move? A.

TANGENT ALERT! Accounting for eye-motion Q. When do we see an object move? A. When its image moves on the retina. Is this really true?

TANGENT ALERT! Accounting for eye-motion (contd. ) The corollary discharge model (Teuber, 1960) Predictions:

TANGENT ALERT! Accounting for eye-motion (contd. ) The corollary discharge model (Teuber, 1960) Predictions: 1. Pushing on the eyeball would cause the world to -------2. A stabilized after-image would appear to ------- when the eye is moved voluntarily 3. If your eye was paralyzed with curare and you then attempted to move it, you would see the world ----

From local motion estimates to global ones: Local motion estimates are ambiguous due to

From local motion estimates to global ones: Local motion estimates are ambiguous due to the ‘Aperture Problem’

Subjective plaids video

Subjective plaids video

From local motion estimates to global ones (contd): Theoretically, the ‘Aperture Problem’ can be

From local motion estimates to global ones (contd): Theoretically, the ‘Aperture Problem’ can be overcome by pooling information across multiple contours.

TANGENT ALERT What happens when no correspondence isTANGENT ALERT What happens when no correspondence is
Tangent Lines Understanding Tangent Lines A tangent toTangent Lines Understanding Tangent Lines A tangent to
BUSINESS CORRESPONDENCE Business Correspondence Business correspondence is anBUSINESS CORRESPONDENCE Business Correspondence Business correspondence is an
INSIGNIFICANCE HAPPENS SUFFERING HAPPENS INSIGNIFICANCE HAPPENS Luke 15INSIGNIFICANCE HAPPENS SUFFERING HAPPENS INSIGNIFICANCE HAPPENS Luke 15
10 1 Tangent Ratios Obj Use tangent to10 1 Tangent Ratios Obj Use tangent to
Tangent and Cotangent Graphs Reading and Drawing TangentTangent and Cotangent Graphs Reading and Drawing Tangent
2 72 8 TANGENT LINES DERIVATIVES TANGENT LINE2 72 8 TANGENT LINES DERIVATIVES TANGENT LINE
Tangent lines n Recall tangent line is theTangent lines n Recall tangent line is the
11 1 Tangent Lines Chapter 11 Circles Tangent11 1 Tangent Lines Chapter 11 Circles Tangent
The Tangent Ratio The Tangent using Angle TheThe Tangent Ratio The Tangent using Angle The
Tangent Line Approximations A tangent line approximation isTangent Line Approximations A tangent line approximation is
The Tangent and Velocity Problems The Tangent ProblemThe Tangent and Velocity Problems The Tangent Problem
13 4 Tangent Ratio 13 4 Tangent Ratio13 4 Tangent Ratio 13 4 Tangent Ratio
The Slope of the Tangent Line The TangentThe Slope of the Tangent Line The Tangent
Tangent Line Recall from geometry Tangent is aTangent Line Recall from geometry Tangent is a
14 4 Tangent Planes Differential Elements Tangent Planes14 4 Tangent Planes Differential Elements Tangent Planes
CORRESPONDENCE CORRESPONDENCE This module is to be usedCORRESPONDENCE CORRESPONDENCE This module is to be used
CORRESPONDENCE Introduction structures in correspondence Any form ofCORRESPONDENCE Introduction structures in correspondence Any form of
Correspondence The correspondence is one of the formsCorrespondence The correspondence is one of the forms