Seite 1 von 1

Exercise 3, Problem 1

Verfasst: 7. Jun 2011 17:37
von tanne
are the example pictures created by the same parameters as we should use, i.e. should our solution look exactly the same when we do it right?

Re: Exercise 3, Problem 1

Verfasst: 7. Jun 2011 19:13
von qgao
tanne hat geschrieben:are the example pictures created by the same parameters as we should use, i.e. should our solution look exactly the same when we do it right?
The example figures are the results of my quick implementation, but I am not sure they are correct. These figures are just used to let you know how to mark the interest points.

Of course the grading would be based on the correctness of your solution, not the similarity to these figures. :)

Re: Exercise 3, Problem 1

Verfasst: 7. Jun 2011 19:15
von tanne
okay, because i get a lot more points with that treshold.

Re: Exercise 3, Problem 1

Verfasst: 10. Jun 2011 10:49
von banshee
me2, seems like a good sign then :)

Re: Exercise 3, Problem 1

Verfasst: 10. Jun 2011 13:47
von DVancura
Hm, I get almost exactly the same result as on the image ...

As far as I can see I only get ~2 points more on the borders where I think I chose another way of working with the pixels outside of the image ('replicate') in my filter operations.

Unfortunately task 2 of problem 1 still looks different :-(

Re: Exercise 3, Problem 1

Verfasst: 11. Jun 2011 16:58
von robert.n
Well, how do we use plot to mark the interest points?

//EDIT:
Use the "hold on" command.

Re: Exercise 3, Problem 1

Verfasst: 11. Jun 2011 18:26
von redDragonfly
i have the same results as DVancura: part one looking almost the same as the assignment paper, while part two looks completely different.
anybody else having the same problem and an idea what the problem might be?

Re: Exercise 3, Problem 1

Verfasst: 13. Jun 2011 16:03
von robert.n
My part 1 and part 2 look almost 100% like the images in the assignment. The only difference is that I detect several additional points at the border of the image.

Re: Exercise 3, Problem 1

Verfasst: 16. Jun 2011 13:34
von redDragonfly
can you give some hints on how you maximized the laplacian? i used the formula on slide 69, like givn on the assignment paper, but i obviously made a mistake somewhere...