Statistical Image & Video Processing HW3: Hidden Markov & Hand Silhouette Model - Prof. Pi, Assignments of Electrical and Electronics Engineering

Two problems from a university-level statistics course focused on image and video processing. The first problem deals with evaluating the probability of observing a sequence of symbols given a hidden markov model. The second problem involves a simplified graphical model for hand silhouettes, where the orientation and angles of the fingers are modeled as random variables. Hints and suggests writing a computer program to evaluate the expected value of the angle of the third finger.

Typology: Assignments

Pre 2010

Uploaded on 03/10/2009

koofers-user-40t
koofers-user-40t 🇺🇸

10 documents

1 / 1

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
ECE544
Statistical Image and Video Processing
Homework #3
Due October 15, 2008
Problem 1 (Hidden Markov Model). Consider the following HMM. The hidden sequence
S0, S1, S2, S3is a homogeneous Markov chain with alphabet Λ = {0,1}and transition prob-
ability matrix Q=0.9 0.1
0.1 0.9. The observed variables are X={X0,·· · , X7} Λ8. The
conditional pmf for Xgiven Sis given by
p(x|s) =
3
Y
i=0
Q(x2i|si)Q(x2i+1|si).
Give a computationally efficient algorithm for evaluating p(x7|s0).
Problem 2 (Simplified graphical model for hand silhouettes). Say the global orientation of
the hand is represented by a variable O, all fingers have prespecified lengths, and the angle
subtended by fingers 1 and 5 (the pinkie and the thumb) is denoted by A. The orientation
of the five fingers is represented by angular variables F1,· · · , F5. In this simplified model, the
hand is parameterized by the 7 variables F1,·· · , F5,O, A.
These variables are modeled as random with the following joint pdf:
p(f1,· · · , f5, o, a) = 1
Zψ15(f1, f5|o, a)ψ12(f1, f2)ψ23 (f2, f3)ψ34(f3, f4)ψ45(f4, f5)p(o)p(a).
1. Draw a graphical representation of this model, and give a computationally efficient algo-
rithm for evaluating p(f3).
2. Does this model appear physiologically reasonable?
Hint: to think about this question, lay one hand flat on a table, hold the pinkie and the
thumb firm with the other hand, and try to move the three free fingers one at a time.
3. Assume now that O=oand A= 10 (degrees) with probability 1, and that
ψi,i+1(fi, fi+1) = exp{fifi+1 }1{fi<fi+1}:i= 1,2,3
= exp{1
4(fifi+1)}1{fi<fi+1}:i= 4
ψ15(f1, f5|o,10) = exp{106|f1+ 5|+ 106|f55|}.
Write a computer program and evaluate the expected value of F3.
Hint: you should be able to make a rough guess of that value before writing the program.
1

Partial preview of the text

Download Statistical Image & Video Processing HW3: Hidden Markov & Hand Silhouette Model - Prof. Pi and more Assignments Electrical and Electronics Engineering in PDF only on Docsity!

ECE

Statistical Image and Video Processing

Homework

Due October 15, 2008

Problem 1 (Hidden Markov Model). Consider the following HMM. The hidden sequence S 0 , S 1 , S 2 , S 3 is a homogeneous Markov chain with alphabet Λ = { 0 , 1 } and transition prob-

ability matrix Q =

. The observed variables are X = {X 0 , · · · , X 7 } ∈ Λ^8. The

conditional pmf for X given S is given by

p(x|s) =

∏^3

i=

Q(x 2 i|si)Q(x 2 i+1|si).

Give a computationally efficient algorithm for evaluating p(x 7 |s 0 ).

Problem 2 (Simplified graphical model for hand silhouettes). Say the global orientation of the hand is represented by a variable O, all fingers have prespecified lengths, and the angle subtended by fingers 1 and 5 (the pinkie and the thumb) is denoted by A. The orientation of the five fingers is represented by angular variables F 1 , · · · , F 5. In this simplified model, the hand is parameterized by the 7 variables F 1 , · · · , F 5 , O, A.

These variables are modeled as random with the following joint pdf:

p(f 1 , · · · , f 5 , o, a) =

Z

ψ 15 (f 1 , f 5 |o, a)ψ 12 (f 1 , f 2 )ψ 23 (f 2 , f 3 )ψ 34 (f 3 , f 4 )ψ 45 (f 4 , f 5 )p(o)p(a).

  1. Draw a graphical representation of this model, and give a computationally efficient algo- rithm for evaluating p(f 3 ).
  2. Does this model appear physiologically reasonable? Hint: to think about this question, lay one hand flat on a table, hold the pinkie and the thumb firm with the other hand, and try to move the three free fingers one at a time.
  3. Assume now that O = o∗^ and A = 10 (degrees) with probability 1, and that

ψi,i+1(fi, fi+1) = exp{fi − fi+1} (^1) {fi<fi+1} : i = 1, 2 , 3

= exp{

(fi − fi+1)} (^1) {fi<fi+1} : i = 4

ψ 15 (f 1 , f 5 |o∗, 10) = exp{ 106 |f 1 + 5| + 10^6 |f 5 − 5 |}.

Write a computer program and evaluate the expected value of F 3. Hint: you should be able to make a rough guess of that value before writing the program.