bikong2 / orb_slam2_dense Goto Github PK

Hi, thank you for making this code available. I have a quick question... Is the PCL dense cloud just grabbed for visualization? Or is the tracking actually done on the dense cloud in this version?

cmake is ok， and make is ok.
when I run the mono_tum , the segmentation fault occurred

Hi, and thank you for making this code available. I am having problems compiling, and get an error:

LNK2019 unresolved external symbol "protected: static struct g2o::G2OBatchStatistics * g2o::G2OBatchStatistics::_globalStats" (?_globalStats@G2OBatchStatistics@g2o@@1PEAU12@EA) referenced in function "public: virtual bool __cdecl g2o::BlockSolver<struct g2o::BlockSolverTraits<6,3> >::computeMarginals(class g2o::SparseBlockMatrix<class Eigen::Matrix<double,-1,-1,0,-1,-1> > &,class std::vector<struct std::pair<int,int>,class std::allocator<struct std::pair<int,int> > > const &)" (?computeMarginals@?$BlockSolver@U?$BlockSolverTraits@$05$02@g2o@@@g2o@@UEAA_NAEAV?$SparseBlockMatrix@V?$Matrix@N$0?0$0?0$0A@$0?0$0?0@Eigen@@@2@AEBV?$vector@U?$pair@HH@std@@V?$allocator@U?$pair@HH@std@@@2@@std@@@Z) ORB_SLAM2 D:\ORB_SLAM2_dense-master\build\Optimizer.obj

I have tried the g2o version that works with vanilla ORBSLAM2, and the one that is included here. What could i be missing?

Thanks again.

this command: $ ./rgbd_tum ../../Vocabulary/ORBvoc.txt TUM1.yaml ~/Downloads/rgbd_dataset_freiburg1_desk ../RGB-D/associations/fr1_desk.txt leads to Segmentation Fault once it tries to build the point cloud for the first keyframe.

Running stereo_euroc

receive a keyframe, id = 2
generate point cloud for kf 2, size=0
show global map, size=0
receive a keyframe, id = 3
generate point cloud for kf 3, size=0
show global map, size=0
receive a keyframe, id = 4
generate point cloud for kf 4, size=0
show global map, size=0
receive a keyframe, id = 5
generate point cloud for kf 5, size=0
show global map, size=0
receive a keyframe, id = 6
generate point cloud for kf 6, size=0
show global map, size=0
receive a keyframe, id = 7
generate point cloud for kf 7, size=0
show global map, size=0
receive a keyframe, id = 8
generate point cloud for kf 8, size=0
show global map, size=0