43
Computer Vision : CISC 4/689 Adaptation from: Prof. James M. Rehg, G.Tech

Adaptation from:

Embed Size (px)

DESCRIPTION

Adaptation from:. Prof. James M. Rehg, G.Tech. Public Library, Stereoscopic Looking Room, Chicago, by Phillips, 1923. Teesta suspension bridge-Darjeeling, India. Mark Twain at Pool Table", no date, UCR Museum of Photography. - PowerPoint PPT Presentation

Citation preview

Page 1: Adaptation from:

Computer Vision : CISC 4/689

Adaptation from:

Prof. James M. Rehg, G.Tech

Page 2: Adaptation from:

Computer Vision : CISC 4/689Public Library, Stereoscopic Looking Room, Chicago, by Phillips, 1923

Page 3: Adaptation from:

Computer Vision : CISC 4/689Teesta suspension bridge-Darjeeling, India

Page 4: Adaptation from:

Computer Vision : CISC 4/689Mark Twain at Pool Table", no date, UCR Museum of Photography

Page 5: Adaptation from:

Computer Vision : CISC 4/689Woman getting eye exam during immigration procedure at Ellis

Island, c. 1905 - 1920 , UCR Museum of Phography

Page 6: Adaptation from:

Computer Vision : CISC 4/689

Stereo results

Ground truthScene

– Data from University of Tsukuba

(Seitz)

Page 7: Adaptation from:

Computer Vision : CISC 4/689

Results with window correlation

Window-based matching(best window size)

Ground truth

(Seitz)

Page 8: Adaptation from:

Computer Vision : CISC 4/689

Results with better method

State of the art methodBoykov et al., Fast Approximate Energy Minimization via Graph Cuts,

International Conference on Computer Vision, September 1999.

Ground truth

(Seitz)

Page 9: Adaptation from:

Computer Vision : CISC 4/689

Stereo Constraint

• Color constancy– The color of any world points remains constant from image to

image– This assumption is true under Lambertian Model– In practice, given photometric camera calibration and typical

scenes, color constancy holds well enough for most stereo algorithms.

Page 10: Adaptation from:

Computer Vision : CISC 4/689

Stereo Constraint

• Epipolar geometry– The epipolar geometry is the fundamental constraint in stereo.

– Rectification aligns epipolar lines with scanlines

Epipolar plane

Epipolar line for pEpipolar line for p’

Page 11: Adaptation from:

Computer Vision : CISC 4/689

Stereo Constraint

• Uniqueness and Continuity– Proposed by Marr&Poggio.

– Each item from each image may be assigned at most one disparity value,” and the “disparity” varies smoothly almost everywhere.

Page 12: Adaptation from:

Computer Vision : CISC 4/689

Problems with Window-based matching

• Disparity within the window must be constant.

• Bias the results towards frontal-parallel surfaces.

• Blur across depth discontinuities.

• Perform poorly in textureless regions.

• Erroneous results in occluded regions

Page 13: Adaptation from:

Computer Vision : CISC 4/689

Cooperative Stereo Algorithm

• Based on two basic assumption by Marr and Poggio:– Uniqueness: at most a single unique match exists for each pixel.

– Continuous: disparity values are generally continuous, i.e., smooth within a local neighborhood.

Page 14: Adaptation from:

Computer Vision : CISC 4/689

Disparity Space Image (DSI)

• The 3D disparity space has dimensions row r column c and disparity d. Each element (r, c, d) of the disparity space projects to the pixel (r, c) in the left image and to the (r, c + d) in the right image

• DSI represents the confidence or likelihood of a particular match.

Page 15: Adaptation from:

Computer Vision : CISC 4/689

Illustration of DSI

(r, c) slices for different d

(c, d) slice for r = 151

Page 16: Adaptation from:

Computer Vision : CISC 4/689

Definition

( , , )nL r c d

0 ( , , )L r c d

( , , )r c d

( , , )r c d

Match value assigned to element (r, c, d) at iteration n

Initial values computed from SSD or NCC

Inhibition area for element (r, c, d)

Local support area for element (r, c, d)

Page 17: Adaptation from:

Computer Vision : CISC 4/689

Illustration of Inhibitory and Support Regions

If d is disparity for (r,c),d -1 is for (r,c-1)And d+1 for (r,c+1), etc.

Vertical inhibition is self explanatory.Diagonal basically means, r,c+1in left imagecannot map to r,c+d in right image, i.e, it cannot have disparity of d-1. Similarly,r,c-1 in left image cannot map to r,c+dso it cannot have disparity of d+1.This satisfies uniqueness constraint.

Page 18: Adaptation from:

Computer Vision : CISC 4/689

Iterative Updating DSI

1

2

3

4

Questions to ponder:i) What is typical alpha?ii) Do we include inhibitionregion in numerator of Rn?

Page 19: Adaptation from:

Computer Vision : CISC 4/689

Explicit Detection of Occlusion

Identify occlusions by examining the magnitude of the converged values in conjunction with the uniqueness

constrain

Page 20: Adaptation from:

Computer Vision : CISC 4/689

Summary of Cooperative Stereo

• Prepare a 3D array, (r, c, d): (r, c) for each pixel in the reference image and d for the range of disparity.

• Set initial match values using a function of image intensities, such as normalized correlation or SSD.

• Iteratively update match values using (4) until the match values converge.• For each pixel (r, c), find the element (r, c, d) with the maximum match value.• If the maximum match value is higher than a threshold, output the disparity d, otherwise,

declare a occlusion.

0L

nL

Page 21: Adaptation from:

Computer Vision : CISC 4/689

Ordering Constraint

• If an object a is left on an object b in the left image then object a will also appear to the left of object b in the right image

Ordering constraint… …and its failure

Page 22: Adaptation from:

Computer Vision : CISC 4/689

Stereo Correspondences

… …Left scanline Right scanline

Match intensities sequentially between two scanlines

Page 23: Adaptation from:

Computer Vision : CISC 4/689

Stereo Correspondences

… …Left scanline Right scanline

Match

Match

MatchLeft occlusion Right occlusion

Page 24: Adaptation from:

Computer Vision : CISC 4/689

Search Over Correspondences

Three cases:– Sequential – cost of match– Left occluded – cost of no match– Right occluded – cost of no match

Left scanline

Right scanline

Left Occluded Pixels

Right occluded Pixels

Page 25: Adaptation from:

Computer Vision : CISC 4/689

Standard 3-move Dynamic Programming for Stereo

Dynamic programming yields the optimal path through grid. This is the best set of matches that satisfy the ordering constraint

Left Occluded Pixels

Left scanline

Right occluded P

ixels

Right scanline

Start

End

Page 26: Adaptation from:

Computer Vision : CISC 4/689

Dynamic Programming• Efficient algorithm for solving sequential decision (optimal path) problems.

1

2

3

1

2

3

1

2

3

1t 2t 3t

1i

2i

3i

1

2

3

Tt

How many paths through this trellis? T3

Page 27: Adaptation from:

Computer Vision : CISC 4/689

Dynamic Programming

1

2

3

1

2

3

1

2

3

1tC tC 1tC

12

22

32

Suppose cost can be decomposed into stages:

jiij state to state from going ofCost

1i

2i

3i

States:

Page 28: Adaptation from:

Computer Vision : CISC 4/689

Dynamic Programming

1

2

3

1

2

3

1

2

3

1tC tC 1tC

12

22

32

Principle of Optimality for an n-stage assignment problem

))((min)( 1 iCjC tijit

2j

1i

2i

3i

Page 29: Adaptation from:

Computer Vision : CISC 4/689

Dynamic Programming

1

2

3

1

2

3

1

2

3

1tC tC 1tC

2)2( tb

))((minarg)(

))((min)(

1

1

iCjb

iCjC

tijit

tijit

2j

1i

2i

3i

Page 30: Adaptation from:

Computer Vision : CISC 4/689

Stereo Matching with Dynamic Programming

Scan across grid computing optimal cost for each node given its upper-left neighbors.Backtrack from the terminal to get the optimal path.

Occluded Pixels

Left scanline

Dis-occluded Pixels

Right scanline

Terminal

Page 31: Adaptation from:

Computer Vision : CISC 4/689

Stereo Matching with Dynamic Programming

Scan across grid computing optimal cost for each node given its upper-left neighbors.Backtrack from the terminal to get the optimal path.

Occluded Pixels

Left scanline

Dis-occluded Pixels

Right scanline

Terminal

Page 32: Adaptation from:

Computer Vision : CISC 4/689

Stereo Matching with Dynamic Programming

Scan across grid computing optimal cost for each node given its upper-left neighbors.Backtrack from the terminal to get the optimal path.

Occluded Pixels

Left scanline

Dis-occluded Pixels

Right scanline

Terminal

Page 33: Adaptation from:

Computer Vision : CISC 4/689

Stereo Matching with Dynamic Programming

Scan across grid computing optimal cost for each node given its upper-left neighbors.Backtrack from the terminal to get the optimal path.

Occluded Pixels

Left scanline

Dis-occluded Pixels

Right scanline

Terminal

Page 34: Adaptation from:

Computer Vision : CISC 4/689

Stereo Matching with Dynamic Programming

Scan across grid computing optimal cost for each node given its upper-left neighbors.Backtrack from the terminal to get the optimal path.

Occluded Pixels

Left scanline

Dis-occluded Pixels

Right scanline

Terminal

Page 35: Adaptation from:

Computer Vision : CISC 4/689

Stereo Matching with Dynamic Programming

Scan across grid computing optimal cost for each node given its upper-left neighbors.Backtrack from the terminal to get the optimal path.

Occluded Pixels

Left scanline

Dis-occluded Pixels

Right scanline

Terminal

Page 36: Adaptation from:

Computer Vision : CISC 4/689

Stereo Matching with Dynamic Programming

Scan across grid computing optimal cost for each node given its upper-left neighbors.Backtrack from the terminal to get the optimal path.

For each (i,j), look for what isOptimal to get to: would it beFrom (i-1,j-1), or (i,j-1), or (i-1,j-1)?Assign large values for left occlusionand right occlusion.

Occluded Pixels

Left scanline

Dis-occluded Pixels

Right scanline

Terminal

Page 37: Adaptation from:

Computer Vision : CISC 4/689

Stereo Matching with Dynamic Programming

Pseudo-code describing how to calculate the optimal match

Page 38: Adaptation from:

Computer Vision : CISC 4/689

Stereo Matching with Dynamic Programming

Pseudo-code describing how to reconstruct the optimal

path

Page 39: Adaptation from:

Computer Vision : CISC 4/689

Results

Local errors may be propagated along a scan-line and no inter scan-line consistency is enforced.

Page 40: Adaptation from:

Computer Vision : CISC 4/689

Assumption Behind Segmentation-based Stereo

• Depth discontinuity tend to correlate well with color edges

• Disparity variation within a segment is small

• Approximating the scene with piece-wise planar surfaces

Page 41: Adaptation from:

Computer Vision : CISC 4/689

Segmentation-based stereo

• Plane equation is fitted in each segment based on initial disparity estimation obtained SSD or Correlation

• Global matching criteria: if a depth map is good, warping the reference image to the other view according to this depth will render an image that matches the real view

• Optimization by iterative neighborhood depth hypothesizing

Page 42: Adaptation from:

Computer Vision : CISC 4/689

Result

Page 43: Adaptation from:

Computer Vision : CISC 4/689

Another Result