A Termination Criterion for Probabilistic PointClouds Registration

TL;DR

This paper introduces a termination criterion for Probabilistic Point Clouds Registration (PPCR) that enables the algorithm to decide when to stop iterating, maintaining accuracy while reducing computational time.

Contribution

The work compares various termination criteria for PPCR and identifies one that balances registration quality and computational efficiency effectively.

Findings

The chosen criterion achieves results comparable to high-iteration runs.

It significantly reduces the number of iterations needed.

The criterion is validated across multiple datasets.

Abstract

Probabilistic Point Clouds Registration (PPCR) is an algorithm that, in its multi-iteration version, outperformed state of the art algorithms for local point clouds registration. However, its performances have been tested using a fixed high number of iterations. To be of practical usefulness, we think that the algorithm should decide by itself when to stop, to avoid an excessive number of iterations and, therefore, wasting computational time. With this work, we compare different termination criterion on several datasets and prove that the chosen one produce very good results that are comparable to those obtained using a very high number of iterations while saving computational time.