orb-slam3

[mesutbulut5]

Hello my teacher. I have homework to do. I aim to find the x and y position in meters on a video using Orb-slam3. Can you help me with this issue? Thanks.

[brmarkus]

Are you referring to one of the samples here under "https://github.com/spmallick/learnopencv" or referring to one of the OpenCV APIs?
There are several repositories around the ORB-SLAM3 paper, with/without IMUs, with/without LiDAR, with/without IntelRealSense-camers. There are different language bindings for the ORB-SLAM3 feature-extraction. With mono and with stereo-images, etc.

What have you tried so far, which point have you reached so far?

[mesutbulut5]

First of all, thank you for your payment. I have a single camera (Monocular) and with this camera I need to estimate the position in meters. I tried some projects on Githup. However, these codes could not be customized. How can I test the problems in my own code? Do you know of any tutorials or blocks on this person? Do you have any work in this field? markusbr ***@***.***>, 1 Haz 2024 Cmt, 14:11 tarihinde şunu yazdı:

…

Are you referring to one of the samples here under " https://github.com/spmallick/learnopencv" or referring to one of the OpenCV APIs? There are several repositories around the ORB-SLAM3 paper, with/without IMUs, with/without LiDAR, with/without IntelRealSense-camers. There are different language bindings for the ORB-SLAM3 feature-extraction. With mono and with stereo-images, etc. What have you tried so far, which point have you reached so far? — Reply to this email directly, view it on GitHub <#918 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ALSCRGPNBC47DDX5SR3HGLDZFGT7HAVCNFSM6AAAAABIUBPKR2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCNBTGQYDSNBWGY> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[brmarkus]

Me personally no experience on ORB-SLAM, sorry.
There are a lot of repositories and for instance examples using Matlab.

If you could provide more details about the exercise, the prerequisites (like model, IMU, camera-calibration, LiDAR, depth-information, ground-truth information? calulcate/estimate the position of what? static or moving, with information about previous frames?) then maybe someone from the community could provide pointers, new hints on the topic.