Coarse Point Cloud Registration Based on Variational Functionals

Point cloud collection forming a 3D scene typically uses information from multiple data scans. The common approach is to register the point pairs consequentially using variant of iterative closest (ICP) algorithm, but most versions ICP algorithm only work correctly for small movement between two clouds. This makes it difficult accumulate Global registration algorithms are also known, which theoretically process clouds at arbitrary initial positions. Recently, multiparameter variational functional was described and used in disadvantage this need manual selection parameters. In paper, modified version with automatic model parameters proposed. proposed fusion RANSAC concepts. Moreover, can be parallelized. performance compared that known global algorithms.