1
$\begingroup$

I have two images and I've found their keypoints using sift keypoint detector, Now I have to match their keypoints with HoG features, I know how to extract HoG description, but I dont know how to combine it with SIFT and match the keypoints, any ideas pls? I am using opencv and python3

$\endgroup$
2
  • $\begingroup$ Would you mind sharing some of your code? $\endgroup$ – wacax Dec 23 '18 at 0:57
  • $\begingroup$ @wacax I have no idea how to code this $\endgroup$ – Saha Dec 23 '18 at 16:56
1
$\begingroup$

In general, you can use brute force or a smart feature matcher implemented in openCV. Another approach is seeing the task as image registration based on extracted features.

The question may be what is the relation of HoG and SIFT if one image has only HoG and other SIFT or both images have detected both features HoG and SIFT.

$\endgroup$
2
  • $\begingroup$ can I detect keypoints using HoG only? I searched a lot and I found out HoG is a descriptor, so I need a detector first. I want to use a sift detector and a HoG descriptor for both images $\endgroup$ – Saha Dec 28 '18 at 17:51
  • $\begingroup$ I would recommend using rather SIFT, SURF or ORB, those are standard options :) $\endgroup$ – Jirka B. Dec 31 '18 at 14:54

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.