Code for the work "Learning Rotations", AGACSE 2021 - MMAS 2022
Following recent literature, we propose three case studies: a sanity check, a pose estimation from 3D point clouds and an inverse kinematic problem. We do so by employing a full geometric algebra (GA) description of rotations. We compare the GA formulation with a 6D continuous representation previously presented in the literature in terms of regression error and reconstruction accuracy. We empirically demonstrate that parametrizing rotations as bivectors outperforms the 6D representation.
The notebooks contain the code to generate the dataset and to process them. The ready to use datasets can be obtained from the authors upon request.