Ask Your Question

warith_h's profile - activity

2015-05-08 13:01:21 -0500 received badge  Good Answer (source)
2014-10-07 02:16:06 -0500 commented answer How to use bag of words example with BRIEF descriptors?
2014-07-17 12:15:24 -0500 commented answer How to use bag of words example with BRIEF descriptors?

My intuitions seem to be confirmed by this link : http://webdiis.unizar.es/~dorian/index.php?p=32 There is a nice implementation with bitset and the mean operator is redefined correctly. One should do something similar in OpenCV.

2014-07-17 11:31:48 -0500 received badge  Necromancer (source)
2014-07-17 11:29:40 -0500 received badge  Nice Answer (source)
2014-07-17 05:13:25 -0500 received badge  Supporter (source)
2014-07-17 04:08:14 -0500 commented answer How to use BRIEF, ORB, FREAK or integer descriptor with cv::BOW?

@Guanta: Thank you. Indeed, we have to look at what Marius Muja did in his implementation of FLANN.

2014-07-17 02:58:19 -0500 received badge  Teacher (source)
2014-07-16 13:13:57 -0500 received badge  Critic (source)
2014-07-16 13:13:43 -0500 answered a question How to use BRIEF, ORB, FREAK or integer descriptor with cv::BOW?

As said in the similar post, this is NOT the right solution, the medoid/mean is not correct giving poor results in my experiments for image retrieval. I am looking for a more correct solution.

2014-07-16 11:28:45 -0500 received badge  Necromancer (source)
2014-07-16 10:50:51 -0500 answered a question How to use bag of words example with BRIEF descriptors?

There is a terrible mistake. A simple conversion of a uchar to float is NOT the right solution. If you look at the details of KMeans, one must minimize the sum of distances :

  • for a real-valued vector, this corresponds to a mean

  • for a binary vector stored in uchar, this does NOT correspond to a mean

Indeed for each coordinate inside the uchar one must make a vote between the ones and the zeros to get a medoid.