Closed JunsukLee closed 3 years ago
It is described here https://github.com/uzh-rpg/rpg_trajectory_evaluation#advanced-different-estimation-type They just look for different file names for the trajectory estimation results, and the analysis afterwards is the same.
It is just a convenient way of feed different estimates into the script, if your algorithm output different types of poses (real-time poses after each image is processed, or bundle adjusted poses after the whole sequence is processed).
Thank you!!
I have a question. What is the difference between vislam-ba and visual-rp? If you can answer my questions, I would be very grateful.