analysis of motion
DESCRIPTION
Analysis of Motion. Measuring motion in biological vision systems. Smoothness assumption:. Compute a velocity field that: is consistent with local measurements of image motion (perpendicular components) has the least amount of variation possible. Computing the smoothest velocity field. - PowerPoint PPT PresentationTRANSCRIPT
![Page 1: Analysis of Motion](https://reader036.vdocuments.site/reader036/viewer/2022062422/56813b43550346895da419c0/html5/thumbnails/1.jpg)
CS332 Visual ProcessingDepartment of Computer ScienceWellesley College
Analysis of Motion
Measuring motion in biological vision
systems
![Page 2: Analysis of Motion](https://reader036.vdocuments.site/reader036/viewer/2022062422/56813b43550346895da419c0/html5/thumbnails/2.jpg)
1-2
Smoothness assumption:
Compute a velocity field that:
(1)is consistent with local measurements of image motion (perpendicular components)
(2)has the least amount of variation possible
![Page 3: Analysis of Motion](https://reader036.vdocuments.site/reader036/viewer/2022062422/56813b43550346895da419c0/html5/thumbnails/3.jpg)
1-3
Computing the smoothest velocity field
(Vxi-1, Vyi-1)
(Vxi, Vyi)
(Vxi+1, Vyi+1)ii+1
i-1
motion components:
Vxiuxi + Vyi uyi= vi
change in velocity:
(Vxi+1-Vxi, Vyi+1-Vyi)
Find (Vxi, Vyi) that minimize:
Σ(Vxiuxi + Vyiuyi - v
i)2 + [(Vxi+1-Vxi)
2 + (Vyi+1-
Vyi)2]
![Page 4: Analysis of Motion](https://reader036.vdocuments.site/reader036/viewer/2022062422/56813b43550346895da419c0/html5/thumbnails/4.jpg)
1-4
When is the smoothest velocity field correct?
When is it wrong?
motion illusions
![Page 5: Analysis of Motion](https://reader036.vdocuments.site/reader036/viewer/2022062422/56813b43550346895da419c0/html5/thumbnails/5.jpg)
1-5
Two-stage motion measurement
motion components 2D image motion
Movshon, Adelson, Gizzi & NewsomeV1: high % of cells selective for direction of motion (especially in layer that projects to MT)
MT: high % of cells selective for direction and speed of motion
lesions in MT behavioral deficits in motion tasks
![Page 6: Analysis of Motion](https://reader036.vdocuments.site/reader036/viewer/2022062422/56813b43550346895da419c0/html5/thumbnails/6.jpg)
1-6
Testing with sine-wave “plaids”
moving plaid
Movshon et al. recorded responses of neurons in area MT to moving plaids with different component gratings
![Page 7: Analysis of Motion](https://reader036.vdocuments.site/reader036/viewer/2022062422/56813b43550346895da419c0/html5/thumbnails/7.jpg)
1-7
The logic behind the experiments…
Component cells measure perpendicular components of motion
e.g. selective for vertical features moving right
predicted responses: (1) (2) (3)
Pattern cells integrate motion components
e.g. selective for rightward motion of pattern
predicted responses: (1) (2) (3)
(1) (2) (3)
![Page 8: Analysis of Motion](https://reader036.vdocuments.site/reader036/viewer/2022062422/56813b43550346895da419c0/html5/thumbnails/8.jpg)
1-8
Movshon et al. observations:
Cortical area V1:
all neurons behaved like component cells
Cortical area MT:
layers 4 & 6: component cells
layers 2, 3, 5: pattern cells
Perceptually, two components are not integrated if:
large difference in spatial frequency
large difference in speed
components have different stereo disparity
Evidence for two-stage motion measurement!
![Page 9: Analysis of Motion](https://reader036.vdocuments.site/reader036/viewer/2022062422/56813b43550346895da419c0/html5/thumbnails/9.jpg)
1-9
Integrating motion over the image
integration along contours vs. over 2D areas:
Nakayama & Silverman
true motio
n perceived motion
contour features can be far away
area features must be
close
![Page 10: Analysis of Motion](https://reader036.vdocuments.site/reader036/viewer/2022062422/56813b43550346895da419c0/html5/thumbnails/10.jpg)
1-10
Recovering 3D structure from motion
![Page 11: Analysis of Motion](https://reader036.vdocuments.site/reader036/viewer/2022062422/56813b43550346895da419c0/html5/thumbnails/11.jpg)
1-11
Ambiguity of 3D recovery
birds’ eye views
We need additional constraint to recover 3D
structure uniquely“rigidity constraint”
![Page 12: Analysis of Motion](https://reader036.vdocuments.site/reader036/viewer/2022062422/56813b43550346895da419c0/html5/thumbnails/12.jpg)
1-12
Image projections
Z
X
perspective projection
image
plane
Z
X
orthographic projection
image plane
(X, Y, Z) (X, Y)
3D 2D(X, Y, Z) (X/Z, Y/Z)
3D 2D