koide3
Sorry, I have been apart from g2o for a few years, and I'm not sure the exact way to implement the proposed approach with g2o. But, in my case with...
You are almost correct. You need to `GICPFactor` that is derived from `gtsam::NonlinearFactor`. That factor class should override `gtsam::NonlinearFactor::linearize` to return a `gtsam::HessianFactor`. Because `GICPFactor` involves two pose variables, you...
Yes, the approach you described is correct. Specifically, you need to implement a GICP factor class derived from `gtsam::NonlinearFactor` that linearizes the GICP error function and returns the linearized function...
In my opinion, you need to compute the error **between points** but not **between poses**. Because this is off topic for this repository, please take a look at the following...
> @koide3 The project includes a `FindIridescence.cmake` module, but it currently does not export any CMake targets. Is it ok by you if I let Conan create a CMake target...
If you are using ROS2, it is a common issue unfortunately. ROS2 has several communication-related issues, and they have not been fixed yet. In that case, [the composition mechanism](https://docs.ros.org/en/foxy/Concepts/About-Composition.html) would...
In [RegistrationResult](https://github.com/koide3/small_gicp/blob/master/include/small_gicp/registration/registration_result.hpp#L11), there are two metrics that may be used as an alternative for `getFitnessScore()`. - `num_inliers` represents the number of source points with valid correspondences (i.e., "num of overlapping...
It may stem from data association, which sometimes causes an oscillation of the objective function. A quick fix is to set the maximum optimization iteration to a small number (e.g.,...
I recognize that GPU-based methods have high demand and am thinking of introducing them to small_gicp. The problem is that GPU implementation requires a completely different software design and would...
Sorry for the late response. What is the version of your ubuntu?