Hi, thanks for your oepn source.
I read through your code and I have questions in the EnergyFunctional::getIMUHessian function (in EnergyFunctional.cpp).
J_all.block ( 0,7+i15,9,15 ) += J_imuiJ_reliJ_r_l_i;
What dose J_r_l_i means? I noticed there is no such term in the supplementary material of VIDSO paper.
And in the DSO code, it updated the PRE_worldToCam by:
PRE_worldToCam = SE3::exp(w2c_leftEps()) get_worldToCam_evalPT();
So it seems to me it used the left multiplicative perturbations, but it's opposite in the IMU preintegration. Shouldn't they take the same form(Left multiplication or right multiplication)?
Thank you!
@RonaldSun
Same question here, I'm also wondering how you derived the right to left Jacobian(J_r_l) for SE3?
Is there any reference we can look up to?
Thanks!
Hi, thanks for your oepn source. I read through your code and I have questions in the EnergyFunctional::getIMUHessian function (in EnergyFunctional.cpp). J_all.block ( 0,7+i15,9,15 ) += J_imuiJ_reliJ_r_l_i; What dose J_r_l_i means? I noticed there is no such term in the supplementary material of VIDSO paper. And in the DSO code, it updated the PRE_worldToCam by: PRE_worldToCam = SE3::exp(w2c_leftEps()) get_worldToCam_evalPT(); So it seems to me it used the left multiplicative perturbations, but it's opposite in the IMU preintegration. Shouldn't they take the same form(Left multiplication or right multiplication)? Thank you!