I have been looking for the motion-only optimization for camera-rate state estimation they mentioned in the paper, but I couldn't find it. Anyone knows if it is in the code and where is it?
As I understood, there is a front-end (feature tracker node) which tracks corners with KLT, and sends these features to the back-end (estimator node) which optimizes the full state vector in the sliding window. They work at 10 Hz
I have been looking for the motion-only optimization for camera-rate state estimation they mentioned in the paper, but I couldn't find it. Anyone knows if it is in the code and where is it?
As I understood, there is a front-end (feature tracker node) which tracks corners with KLT, and sends these features to the back-end (estimator node) which optimizes the full state vector in the sliding window. They work at 10 Hz