Ask Your Question

Combine SIFT with other method for object recognition

asked 2012-07-28 05:38:00 -0600

yes123 gravatar image

Basically I need to develop an object recognition system. I have already used with success matching with SIFT or ORB.

Now I would like to use in the same system something else to improve SIFT/ORB, for example shape matching after a canny edge extractor (or whatever)

Can you suggest me a way to proceed?

I was thinking to always start

  • with corner detection (like SIFT,ORB)
  • then compute ORB descriptors for the corners (like standard ORB)
  • now extract some more information around a patch centered in the corner (contour infroamtion?)

Any Hint?

edit retag flag offensive close merge delete


I would recommend you to provide more details in your original question ( Or you can do it here, but there is no point in keeping both of them.

Kirill Kornyakov gravatar imageKirill Kornyakov ( 2012-07-29 00:49:34 -0600 )edit

@Kirill: I believe there are 2 different topic. One thing is to do an object detection with corner information, another thing is to combine different method into one

yes123 gravatar imageyes123 ( 2012-07-29 06:10:44 -0600 )edit

1 answer

Sort by ยป oldest newest most voted

answered 2012-10-29 05:06:23 -0600

Mahdi gravatar image

updated 2012-10-29 05:07:03 -0600

A very powerful method is using HOG detector! You may get a confidence on regions you have detected by a HOG detector and SVM!
Original Paper
Also you may use Bag of words model that is a step after extracting SIFT features.

edit flag offensive delete link more

Question Tools

1 follower


Asked: 2012-07-28 05:38:00 -0600

Seen: 1,654 times

Last updated: Oct 29 '12