In case of image inputs for estimateRigidTransform() what feature points are computed? Are they similar to SIFT/SURF/FAST/ORB?

affinetransform

asked 2017-07-03 15:21:45 -0600

ambikav
1 ●1 ●1 ●1

I'm starting to work with estimateRigidTransform(). It allows two types of inputs - Vector or mat of 2D points OR two input images. It is mentioned in docs that some feature points are computed in case the inputs are images. What are these features?

edit retag flag offensive close merge delete

add a comment

1 answer

Sort by » oldest newest most voted

answered 2017-07-04 01:06:52 -0600

berak
32993 ●7 ●81 ●312

updated 2017-07-04 01:43:42 -0600

if you pass grayscale (CV_8U) images to estimateRigidTransform(), it will use sparse optical flow based on simple grid points, see src code

edit flag offensive delete link

Comments

Thanks for clarifying this!

ambikav ( 2017-07-05 19:22:41 -0600 )edit

add a comment

Links

Question Tools

1 follower

Stats

Asked: 2017-07-03 15:21:45 -0600

Seen: 437 times

Last updated: Jul 04 '17

In case of image inputs for estimateRigidTransform() what feature points are computed? Are they similar to SIFT/SURF/FAST/ORB? edit