iwander-all

why imupreintegration multiply with gnorm ?

2

Hello, I notice that in` void IMUIntegrator::PreIntegration()`, it shows: ``` Eigen::Vector3d acc; //为啥要乘以gnorm？？？？ acc linear_acceleration.x * gnorm, imu->linear_acceleration.y * gnorm, imu->linear_acceleration.z * gnorm; ``` why do so? thanks! 请问大佬，为什么imu与积分这里的加速度要乘以g

So sad to know you this way. Hope the other world is nicer and warmer. R.I.P.

博主，我是你的忠实粉丝，有一个疑惑我困扰了很久，就是在代码里，状态量的增量是solution = K * meas_vec + vec - K * Hsub * vec.block(0, 0)(1)，但是论文fast-lio公式(18)和代码似乎是正负号相反的，如果按照论文，应该是solution =- K * meas_vec - vec + K * Hsub * vec.block(0, 0)(2)，在阅读fast-livo和r3live的代码，发现它们的lio部分都是与(1)一致，而vio部分与(2)一致，请问作者如何理解

error: no match for ‘operator-’ (operand types are ‘ImVec2’ and ‘ImVec2’)

2

hi， when compile `cmake --build build --config RelWithDebInfo` Ubuntu 18, it shows: I did not find it in existing issues, how to solve it, thanks! `[ 3%] Built target optix_program...

I am new user of nerf, could you offer a demo data satisfy your code, thanks

pointcloud saved in g_eigen_vec_vec seems useless and spend lots of memory. But when I stop put point inside, it fails in save to ply meshes. could you offer a help

MinkowskiEngine install

1

` python setup.py install --blas_include_dirs=${CONDA_PREFIX}/include --blas=openblas ` should add ` --force_cuda`

large scale pointCloud

2

Hello , big olds, could your masterpiece handle large scale pointCloud inputs to nice mesh ?

Hi, big olds, may you please offer a comparision between R-VIO 1 and 2 ? Is 2 better ? In what aspect?