-
Hi! I have some questions about the design of Match R-CNN:
1) What are the coefficient lambda_1 ~ lambda_5 in the loss function L?
2) I would like to know the details to create pairs in training sta…
-
Sorry for the long report in advance, I hope developers can react to this, and agree to find a better approach.
The existence of `Feature2D::detect` and `Feature2D::compute` advertises for detector…
ghost updated
2 years ago
-
Hi - thank you for providing the pretrained weights - saves us a lot of time to try this out as an option!
If I understand correctly, SuperGlue operates only on pairs of images (please correct me i…
-
can we use deep learning or an ai model to return only good matches?
Yes, deep learning and AI models can be used to improve feature matching by learning more discriminative feature descriptors or …
-
I would like to use only selected features and their depth values through post-processing rather than features obtained through the frame constructor.
I mean...I want to estimate the motion with on…
-
@targos @stropitek @lpatiny
Since Harris corner detection and FAST corner detection algorithms are two mainstream corner detection solutions, They are critical for processing images in Web systems.…
-
I am trying to get the library working on Ubuntu 14.04. I upgraded my cmake from 2.8 to 3.2.2 (installed via PPA according to this post - https://askubuntu.com/a/610352/133573) but the library will no…
-
Hi, @gfacciol . When I process large images, the log file shows that:
and the termi…
-
**Describe the bug**
> I0718 20:00:47.199163 64287 pairing.cc:318] Matching image [2261/29302]
E0718 20:00:47.207983 64287 database.cc:749] SQLite error: database disk image is malformed
terminat…
-
I am trying to provide my own feature extraction & matching by the commands feature_importer, and matches_importer the files were read successfully but I got the following warnings and the mapper stag…