Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

not limitKps inside Feature2D::generateKeypoints when using pyDetector #1223

Merged
merged 2 commits into from
Feb 26, 2024

Conversation

hellovuong
Copy link
Contributor

This only happens with SuperPoint. Rtabmap will limit key points if exceeds the limit set by Kp/MaxFeatures THEN it will generate descriptors based on extracted key points. It will work classic Features Extraction method. However, SuperPoint will generate key points and descriptors at the same time, if the limit key points method calls inside the extract key point method it will lead to a mismatched number of key points and descriptors later when the generate descriptor method calls. It is solved by ignoring the limit key points method inside the generate key point method and only limiting key points after generating the descriptors if necessary.

@hellovuong
Copy link
Contributor Author

hellovuong commented Feb 26, 2024

Sorry, there is a wrong pull. I will fix it soon! FIXED

@matlabbe
Copy link
Member

matlabbe commented Feb 26, 2024

Good catch. I'll merge it now as it can cause errors, but I opened a new issue #1225 to think about supporting that feature for PyDetector. Without this patch, the workaround would be to set Vis/MaxFeatures=0 and limit the keypoints/descriptors inside the python file. For the visual word dictionary, Kp/MaxFeatures should still work as expected.

@matlabbe matlabbe merged commit 752ca76 into introlab:master Feb 26, 2024
6 checks passed
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.

None yet

2 participants