Question: Question 5 . Patch Matching. ( 2 0 points ) We have discussed different scoring functions for measuring how well two patches of pixels match

Question 5. Patch Matching. (20 points) We have discussed different scoring functions for measuring how well two patches of pixels match each other. Three of these functions were sum of squared differences (SSD), correlation, and normalized cross correlation (NCC). Let two intensity patches be:
900
099
f=
900
g=
09
9
99
9
000
A.(4 pts) Compute the correlation score between patches f and g.
B.(4 pts) Compute the SSD score between patches f and g.
C.(8 pts) Compute the normalized correlation (NCC) score between patches f and g. Recall that NCC is, in general, a floating point number. Enter value rounded off to 2 decimal places
D.(2 pts) Considered as a template to compare against patches in an image, f is suitable for finding EDGES or CORNERS (choose the correct one).
E.(2 pts) Is f a separable filter? That is, can it be generated as the convolution of a row vector with a column vector? (YES or NO).

Step by Step Solution

There are 3 Steps involved in it

1 Expert Approved Answer
Step: 1 Unlock blur-text-image
Question Has Been Solved by an Expert!

Get step-by-step solutions from verified subject matter experts

Step: 2 Unlock
Step: 3 Unlock

Students Have Also Explored These Related Databases Questions!