Open brianm-sra opened 2 years ago
I do get some additional numeric values for EuRoC V102 stereo only and V103 stereo only (such as Loop Fusion, Total Loop Closing, GBA, Total Full GBA)
So it appears to be an issue only for some datasets, not all
I still saw nan for Map Merging. Under what circumstances does Map Merging happen?
When I add #define REGISTER_TIMES in the system header file I get a segmentation fault after the viewer starts. Did you also encounter this problem @brianm-sra ?
ORB-SLAM3 Copyright (C) 2017-2020 Carlos Campos, Richard Elvira, Juan J. Gómez, José M.M. Montiel and Juan D. Tardós, University of Zaragoza. ORB-SLAM2 Copyright (C) 2014-2016 Raúl Mur-Artal, José M.M. Montiel and Juan D. Tardós, University of Zaragoza. This program comes with ABSOLUTELY NO WARRANTY; This is free software, and you are welcome to redistribute it under certain conditions. See LICENSE.txt.
SLAM settings: -Camera 1 parameters (Pinhole): [ 348.522 348.45 344.484 188.808 ] -Camera 2 parameters (Pinhole -Original image size: [ 672 , 376 ] -Current image size: [ 672 , 376 ] -Sequence FPS: 15 -Stereo baseline: 0.118718 -Stereo depth threshold : 41.2728 -Features per image: 1250 -ORB scale factor: 1.2 -ORB number of scales: 8 -Initial FAST threshold: 20 -Min FAST threshold: 7
Loading ORB Vocabulary. This could take a while... Vocabulary loaded!
Initialization of Atlas from scratch Creation of new map with id: 0 Creation of new map with last KF id: 0 Seq. Name: There are 1 cameras in the atlas Camera 0 is pinhole Starting the Viewer Segmentation fault (core dumped)
I uncommented this line in include/Settings.h :
define REGISTER_TIMES
This is the "Flag to activate the measurement of time in each process (track,localmap, place recognition)."
I then ran some EuRoC and TUM-VI data sets
The output for each included a number of "nan" (not a number) results, such as shown in the quoted block below.
Has anyone else seen this? Any idea what the root cause could be?