Closed abscanxcc closed 1 year ago
Were you able to solve this issue?
Hi, thanks for your feedback. I think it is very helpful to consider that:
ORBSLAM
was designed and runs under C++ in a multithreading way,pyslam
was designed and runs under Python, where the GIL does not allow real multithreading (everything runs on a single core)As stated in the main README
_"You can use this framework as a baseline to play with local features, VO techniques and create your own (proof of concept) VO/SLAM pipeline in python. When you test it, consider that's a work in progress, a development framework written in Python, without any pretense of having state-of-the-art localization accuracy or real-time performances".
pyslam
was NOT designed to have real-time performance.
That being said, you can save "current" estimates as explained here https://github.com/luigifreda/pyslam/issues/97#issuecomment-1762851078 A function to save the "final" estimates is currently missing. Work in progress.
As for the EVO format, it's a question of understanding its requirements.
Hope this helps. Luigi
i want to save the estimated trajectory from slam system for evo i tried to save the data like this but i find it is different from the output using ORB_SLAM2 could someone teach me how to save the trajectory in a correct way?
ORB_SLAM2 output:
My output: