Replies: 1 comment
-
Hi,
Yes, exactly :) |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hi,
Yes, exactly :) |
Beta Was this translation helpful? Give feedback.
-
Hi!
I need help with properly using the find_essential method.
I have 2D point correspondences for image pair and I know that from that data I can estimate the fundamental matrix. If I have fundamental matrix, I can multiply it with calibration matrices to get the essential matrix.
In opencv, method for estimating the essential matrix takes also K (calibration) together with corresponding points, but in Kornia it only takes the points (no calibration needed?) and that makes me confused.
I have seen an example (here) where points are normalized before estimating the essential matrix and than for K an eye matrix is given.
Should it be the same here in Kornia?
@ducha-aiki I'm tagging you since I have seen that you have LGTM-ed that merge request integrating this Nister's 5-point alg for the estimation. :)
So, if I have
kpts0
,kpts1
,K
, with shapes:kpts0
: (1, N, 2)kpts1
: (1, N, 2)K
: (1, 3, 3)what do I need to do to properly use the
find_essential
method in Kornia and get the essential matrix?Thanks!
Beta Was this translation helpful? Give feedback.
All reactions