matcher tutorial
hi all...
was following the matching tutorial at: http://docs.opencv.org/doc/tutorials/features2d/feature_homography/feature_homography.html#feature-homography
i pretty much used the same code but the results with my own images are not that great. as you can see on the image bellow the mapped key points are not really as densely located as with the example box images. is there a special procedure to follow to prepare the images prior to using this code? i did use grayscale pngs as the example box.png and box_in_scene.png...
here is my result: http://el.net/img/tmp/result.png
thanks...
(also whenever writing a title for a question here an annoying suggest menu is showing up and no matter what or where i am clicking it does not disappear. it just pushes the text of the message to the right)