A qualitative review on 3D coarse registration methods
3D registration or matching is a crucial step in 3D model reconstruction. Registration applications span along a variety of research fields, including computational geometry, computer vision, and geometric modeling. This variety of applications produces many diverse approaches to the problem but at...
| Autores: | , , , |
|---|---|
| Tipo de recurso: | artículo |
| Estado: | Versión publicada |
| Fecha de publicación: | 2015 |
| País: | España |
| Institución: | Varias* (Consorci de Biblioteques Universitáries de Catalunya, Centre de Serveis Científics i Acadèmics de Catalunya) |
| Repositorio: | Recercat. Dipósit de la Recerca de Catalunya |
| OAI Identifier: | oai:recercat.cat:10256/12497 |
| Acceso en línea: | http://hdl.handle.net/10256/12497 |
| Access Level: | acceso embargado |
| Palabra clave: | Algorismes computacionals Computer algorithms Geometria computacional Computational geometry |
| Sumario: | 3D registration or matching is a crucial step in 3D model reconstruction. Registration applications span along a variety of research fields, including computational geometry, computer vision, and geometric modeling. This variety of applications produces many diverse approaches to the problem but at the same time yields divergent notations and a lack of standardized algorithms and guidelines to classify existing methods. In this article, we review the state of the art of the 3D rigid registration topic (focused on Coarse Matching) and offer qualitative comparison between the most relevant approaches. Furthermore, we propose a pipeline to classify the existing methods and define a standard formal notation, offering a global point of view of the literature. Our discussion, based on the results presented in the analyzed papers, shows how, although certain aspects of the registration process still need to be tested further in real application situations, the registration pipeline as a whole has progressed steadily. As a result of this progress in all registration aspects, it is now possible to put together algorithms that are able to tackle new and challenging problems with unprecedented data sizes and meeting strict precision criteria |
|---|