Stereo Result

[ZzhYgwh]

Thanks for the author and this open source work！
But now I am using the EuRoC dataset to reproduce the Stereo experiment, and the results of the evo evaluation make me a little confused, even though I used -as to align and correct the scale. Some reference pictures are as follows:
MH05

and V202.

I run this code as:
./bin/HyperSLAM resources/datasets/euroc/setups/stereo/settings.yaml output/euroc/
and modify the output file as TUM format
This seems to be completely unrelated to the truth trajectory. Is there anyone who can answer this question? Thank you very much!

[McEmzi]

@ZzhYgwh Thank you for your question. I am aware that some people raised questions/issues (see #11) related to the stereo-inertial operation, which was left to be added in a later patch. However, the stereo-only version seemed to be working for others. You mentioned some conversions to TUM format... Might it be possible that some columns of the position and/or quaternions have a different order than expected?