Skip to content

Conversation

@Skwangles
Copy link
Contributor

Landmarks are made from the size of undistorted keypoints, so they are exactly the same size.
If frame_publisher::update() is called in between get_keypoints() and get_landmarks() this assertion may not hold.

Add function which returns both in a pair without releasing the mutex lock


img.copyTo(img_);

assert(keypts.size() == curr_lms.size());
Copy link
Contributor Author

@Skwangles Skwangles Jan 6, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I haven't seen anywhere in the frame_publisher that asserts this assumption, and it is not clear to me as a reader the relationship between keypts and curr_lms, so I thought i'd add one.

With the issue with iridescence viewer having these two mismatched, it just feels right to assert the assumption here to make clear their relationship and track any issues further up the chain.

@ymd-stella ymd-stella self-requested a review January 6, 2025 11:42
@ymd-stella ymd-stella merged commit 1d6a595 into stella-cv:main Jan 6, 2025
5 checks passed
@Skwangles Skwangles deleted the keypoint-landmarks branch January 6, 2025 21:52
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants