The uncertainty propagation library is a collection of codes that allow for the estimation and propagation of uncertainties in the solutions of minimal solvers.
The library is build on top a general scheme for propagating uncertainty published in paper:
- Barath, D., Polic, M., Förstner, W., Sattler, T., Pajdla, T., & Kukelova, Z. (2020). Making affine correspondences work in camera geometry computation. In Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part XI 16 (pp. 723-740). Springer International Publishing.
To run the code, please follow these steps:
- Check the "uncertainty_propagation_test.m" file to verify that the propagation works correctly.
- Navigate to the directory of the specific minimal problem you want to run, such as "relative_pose."
- At the beginning of each uncertainty propagation code, you will find a description of the inputs and outputs.
- To see a practical example of the input data to use, refer to the tests provided.
Following these steps will help ensure a successful execution of the code.