Click here to Skip to main content
Rate this: bad
good
Please Sign up or sign in to vote.
See more: C#3.0
Hi friend, currently i am working on a project in which i have to use SIFT(Scale invariant future transform). Actually I want to develop an application like search engine in that i want to get know user intension when he click on any image in web browser. when he click on any image i want to show similar types of images from my database on browser.In this process i have to strictly implement SIFT algorithm.
I have SIFT project but when i run that SIFT project & give my images as input and change keyPoint combobox & select as 1300 or "double input size" or gives any large number in all case it does not gives me more than 1028 kaepoints in output xml file which contains x,y points etc. & when i try to change something in SIFT code it does not allows me to change any thing as it is read only (as i have added library files like libsift). So can any one help me to how to get more point of interest means 3000 or 4000 x & y points in xml file. Thank you.
Posted 14-Nov-12 7:56am

1 solution

Rate this: bad
good
Please Sign up or sign in to vote.

Solution 1

It seems to me that more points of interest come from more data. How else could it work ? You've posted a lot of text, and no code, it's pretty hard to follow, but if you want help tweaking your algorithm b.c you think you have more meaningful data and you're not finding it, we'd need examples, and code. If you just want more hits, you need more data.
  Permalink  
Comments
maheshpardeshi at 14-Nov-12 14:19pm
   
thanks Christian for replay,as there is lots of classes in project & lots of code so not possible to paste code here you can download it from http://www.nowozin.net/sebastian/tu-berlin-2006/autopano-sift/ & i have to give my fix images as input so those i will give any image as input it should give me more points as Point of interest.actually there are lacks of pixels in my image but it is giving only 1000 as point of interest so it is difficult to detect the same object in different image.
Christian Graus at 14-Nov-12 14:20pm
   
I'm not sure I follow all of that, but if you can't post some code and some examples of data you expect to hit that misses, I'm not sure how anyone can help you

This content, along with any associated source code and files, is licensed under The Code Project Open License (CPOL)

  Print Answers RSS
0 OriginalGriff 248
1 Maciej Los 210
2 Sergey Alexandrovich Kryukov 160
3 Peter Leow 155
4 George Jonsson 135


Advertise | Privacy | Mobile
Web03 | 2.8.140709.1 | Last Updated 14 Nov 2012
Copyright © CodeProject, 1999-2014
All Rights Reserved. Terms of Service
Layout: fixed | fluid