Parskatt
Ask for advice about merging match_keypoints into hloc sfm pipeline
I've tried using superpoint and also dedode as keypoints extractor and then match with RoMa but can't get a proper result. How could I improve the result? The basic idea is that I firstly extract all keypoints and then normalize the keypoints into range [-1, 1].
def to_reference_coord(kpt, W, H): kpts = torch.stack((kpt[..., 0]/(W/2) -1, kpt[..., 1]/(H/2) -1), axis=-1) return kpts
And then using roma model to calculate the warp and certainty. then using kpts, warp and certainty as input of match_keypoints. After getting the results I also used ransac to filter some of the matches.
I'm not able to reconstruct the model correctly by this way.
Thus I check the visual result of matches:
Anybody could give me some advice to improve the pipeline? Thanks!
By the way, this magic pickup truck is really hard to reconstruct. LOL
I would avoid matching when the roma confidence is very low. In the provided images I would guess that is the case.
I would avoid matching when the roma confidence is very low. In the provided images I would guess that is the case.
Actually RoMa reaches pretty accurate result by RegressionMatcher.match. Thank you for your reply, I will firstly have a review on my code.
I've made an updated version of the keypoint matching, which also takes the distance from the predicted warp into account. Also I've set the conf threshold by default to 0. This seems to work quite well in my internal tests.
Available from the latest commit.
Has your problem been solved, is this approach feasible? Thank you!
I'm recently working on many other tasks, but the author has updated the code in the most recent commit, you could try to have a check.
I've made an updated version of the keypoint matching, which also takes the distance from the predicted warp into account. Also I've set the conf threshold by default to 0. This seems to work quite well in my internal tests.
Available from the latest commit.
thank you so much
@DabblerGISer I'm doing the same thing：https://github.com/Parskatt/RoMa/issues/102 , I think my steps are exactly right, but matching the dedode keypoints by roma is almost all wrong, did you solve it
I think this should be solved by the update. @Bin-ze open a new issue if still having issues.