Multi-view registration of unordered range scans by fast correspondence propagation of multi-scale descriptors

TL;DR

This paper introduces a fast, reliable method for multi-view registration of unordered range scans by enhancing pair-wise registration with descriptor propagation, reliability judgment, and model augmentation, leading to accurate results.

Contribution

It presents a novel global approach combining descriptor propagation, reliability assessment, and model augmentation for efficient multi-view registration.

Findings

Achieves automatic multi-view registration with high accuracy

Outperforms existing methods on public datasets

Demonstrates robustness and efficiency in registration process

Abstract

This paper proposes a global approach for the multi-view registration of unordered range scans. As the basis of multi-view registration, pair-wise registration is very pivotal. Therefore, we first select a good descriptor and accelerate its correspondence propagation for the pair-wise registration. Then, we design an effective rule to judge the reliability of pair-wise registration results. Subsequently, we propose a model augmentation method, which can utilize reliable results of pair-wise registration to augment the model shape. Finally, multi-view registration can be accomplished by operating the pair-wise registration and judgment, and model augmentation alternately. Experimental results on public available data sets show, that this approach can automatically achieve the multi-view registration of unordered range scans with good accuracy and effectiveness.