Computer Vision - A Modern Approach
Set: Tracking
Slides by D.A. Forsyth
The three main issues in tracking
Computer Vision - A Modern Approach
Set: Tracking
Slides by D.A. Forsyth
Tracking
Very general model:
We assume there are moving objects, which have an underlying state X
There are measurements Y, some of which are functions of this state
There is a clock
at each tick, the state changes
at each tick, we get a new observation
Examples
object is ball, state is 3D position+velocity, measurements are stereo pairs
object is person, state is body configuration, measurements are frames,
clock is in camera (30 fps)
Computer Vision - A Modern Approach
Set: Tracking
Slides by D.A. Forsyth
Three main steps
Computer Vision - A Modern Approach
Set: Tracking
Slides by D.A. Forsyth
Simplifying Assumptions
Computer Vision - A Modern Approach
Set: Tracking
Slides by D.A. Forsyth
Tracking as induction
Assume data association is done
well talk about this later; a dangerous assumption
Do correction for the 0th frame
Assume we have corrected estimate for ith frame
show we can do prediction for i+1, correction for i+1
Computer Vision - A Modern Approach
Set: Tracking
Slides by D.A. Forsyth
Base case
Computer Vision - A Modern Approach
Set: Tracking
Slides by D.A. Forsyth
Induction step
Given
Computer Vision - A Modern Approach
Set: Tracking
Slides by D.A. Forsyth
Induction step
Computer Vision - A Modern Approach
Set: Tracking
Slides by D.A. Forsyth
Linear dynamic models
Use notation ~ to mean has the
pdf of, N(a, b) is a normal
distribution with mean a and
covariance b.
Then a linear dynamic model
has the form
This is much, much more
general than it looks, and
extremely powerful
y
i
= N M
i
x
i
; E
m
i
( )
x
i
= N D
i1
x
i1
; E
d
i
( )
Computer Vision - A Modern Approach
Set: Tracking
Slides by D.A. Forsyth
Examples
Drifting points
we assume that the new position of the point is the old one, plus
noise.
For the measurement model, we may not need to observe the
whole state of the object
e.g. a point moving in 3D, at the 3kth tick we see x, 3k+1th
tick we see y, 3k+2th tick we see z
in this case, we can still make decent estimates of all three
coordinates at each tick.
This property, which does not apply to every model, is called
Observability
Computer Vision - A Modern Approach
Set: Tracking
Slides by D.A. Forsyth
Examples
Points moving with constant velocity
Periodic motion
Etc.
Points moving with constant acceleration
Computer Vision - A Modern Approach
Set: Tracking
Slides by D.A. Forsyth
Points moving with constant velocity
We have
(the Greek letters denote noise terms)
Stack (u, v) into a single state vector
which is the form we had above
u
i
= u
i1
+ Atv
i1
+ c
i
v
i
= v
i1
+
i
u
v
|
\
|
.
|
i
=
1 At
0 1
|
\
|
.
|
u
v
|
\
|
.
|
i1
+ noise
Computer Vision - A Modern Approach
Set: Tracking
Slides by D.A. Forsyth
Points moving with constant acceleration
We have
(the Greek letters denote noise terms)
Stack (u, v) into a single state vector
which is the form we had above
u
i
= u
i1
+ Atv
i1
+ c
i
v
i
= v
i1
+ Ata
i1
+
i
a
i
= a
i1
+
i
u
v
a
|
\
|
.
|
|
i
=
1 At 0
0 1 At
0 0 1
|
\
|
.
|
|
u
v
a
|
\
|
.
|
|
i1
+ noise
Computer Vision - A Modern Approach
Set: Tracking
Slides by D.A. Forsyth
Computer Vision - A Modern Approach
Set: Tracking
Slides by D.A. Forsyth
Computer Vision - A Modern Approach
Set: Tracking
Slides by D.A. Forsyth
The Kalman Filter
Key ideas:
Linear models interact uniquely well with Gaussian noise - make
the prior Gaussian, everything else Gaussian and the calculations
are easy
Gaussians are really easy to represent --- once you know the mean
and covariance, youre done
Computer Vision - A Modern Approach
Set: Tracking
Slides by D.A. Forsyth
The Kalman Filter in 1D
Dynamic Model
Notation
Predicted mean
Corrected mean
Computer Vision - A Modern Approach
Set: Tracking
Slides by D.A. Forsyth
Prediction for 1D Kalman filter
The new state is obtained by
multiplying old state by known constant
adding zero-mean noise
Therefore, predicted mean for new state is
constant times mean for old state
Predicted variance is
sum of constant^2 times old state variance and noise variance
Because:
old state is normal random variable, multiplying normal rv by constant
implies mean is multiplied by a constant variance by square of constant, adding
zero mean noise adds zero to the mean, adding rvs adds variance
Computer Vision - A Modern Approach
Set: Tracking
Slides by D.A. Forsyth
Computer Vision - A Modern Approach
Set: Tracking
Slides by D.A. Forsyth
Correction for 1D Kalman filter
Pattern match to identities given in book
basically, guess the integrals, get:
Notice:
if measurement noise is small,
we rely mainly on the measurement,
if its large, mainly on the
prediction
Computer Vision - A Modern Approach
Set: Tracking
Slides by D.A. Forsyth
In higher dimensions,
derivation follows the
same lines, but isnt as
easy. Expressions here.
Computer Vision - A Modern Approach
Set: Tracking
Slides by D.A. Forsyth
Computer Vision - A Modern Approach
Set: Tracking
Slides by D.A. Forsyth
Computer Vision - A Modern Approach
Set: Tracking
Slides by D.A. Forsyth
Smoothing
Idea
We dont have the best estimate of state - what about the future?
Run two filters, one moving forward, the other backward in time.
Now combine state estimates
The crucial point here is that we can obtain a smoothed
estimate by viewing the backward filters prediction as yet
another measurement for the forward filter
so weve already done the equations
Computer Vision - A Modern Approach
Set: Tracking
Slides by D.A. Forsyth
Computer Vision - A Modern Approach
Set: Tracking
Slides by D.A. Forsyth
Computer Vision - A Modern Approach
Set: Tracking
Slides by D.A. Forsyth
Computer Vision - A Modern Approach
Set: Tracking
Slides by D.A. Forsyth
Data Association
Nearest neighbours
choose the measurement with highest probability given predicted
state
popular, but can lead to catastrophe
Probabilistic Data Association
combine measurements, weighting by probability given predicted
state
gate using predicted state
Computer Vision - A Modern Approach
Set: Tracking
Slides by D.A. Forsyth
Computer Vision - A Modern Approach
Set: Tracking
Slides by D.A. Forsyth
Computer Vision - A Modern Approach
Set: Tracking
Slides by D.A. Forsyth
Computer Vision - A Modern Approach
Set: Tracking
Slides by D.A. Forsyth
Computer Vision - A Modern Approach
Set: Tracking
Slides by D.A. Forsyth
Computer Vision - A Modern Approach
Set: Tracking
Slides by D.A. Forsyth
