How to construct descriptors for MSER and then do the matching?

asked 2016-11-08 05:32:12 -0600

ElectronicEng2015 gravatar image

updated 2016-11-08 05:36:02 -0600

According to what I have read, MSER feature detector identifies homogeneous stable regions in an image. To each of those regions it is possible to fit ellipses to them, so that I can identify the ellipses orientation respect to the vertical axis of the image, then perform affine transformation to rotate the image accordingly, extract the patch containing the ellipse and then create a symmetrical fixed size blob ( 30 pixeles x 30 pixeles, in my case).

From there on, I need to compute the descriptors of each of the blobs (affine normalization) and then compute the matching algorithm.

The question is how do I build up the descriptors from the blobs mentioned above? which matcher is it better to use for those descriptors? And finally, is it possible to simply take the center of the original fitted ellipses as the keypoints?

I thank in advance any help you can provide me

Best regards

edit retag flag offensive close merge delete


It is quite unclear to me why you would like to transform the found ellipses fitted to MSER regions back to a circular shape. What you are doing then is transforming your original data related to stableness but not related to any deformation. This will generate data that is unnatural. Isn't it better to create a feature descriptor on top of the ellipses?

StevenPuttemans gravatar imageStevenPuttemans ( 2016-11-08 09:01:59 -0600 )edit

Hi Steven, I just read that after fitting the ellipses, as they present an orientation angle, it could be used to improve the affine invariance of the image that is being tracked. Now what I don't really understand is how to create the feature descriptor from the fitted ellipse, because I know I can extract the bounding box around that ellipse as a patch and then resize it to a fixed size image. I would appreciate if you can describe the method for feature descriptor creation. Thanks in advance

ElectronicEng2015 gravatar imageElectronicEng2015 ( 2016-11-08 22:07:40 -0600 )edit

Do not forget, that it states, it could be used. This means that if your point of interest does not have an affine deformation, then it is of no use. Calculating the descriptor depends, MSER gives you regions, you now have to decide which type of feature vector you want extracted from that as descriptor. MSER does not supply that part.

StevenPuttemans gravatar imageStevenPuttemans ( 2016-11-09 03:49:14 -0600 )edit