Dear Authors,
thank you very much for your dataset!
I'm trying to run Spagna with visual-inertial SLAM, but couldn't make it work properly without online intrinsic (K matrix and distortion coefficients) calibration. Actually, it turns out that focal lengths and principal points converged to somewhat different values than you provided.
May I ask whether it is possible for you to provide calibration sequences so that I can calibrate myself?
Best regards,
Jaehyung