Geometric Primitive Fitting in Large Structure 3D Point Clouds Acquired by Drones

  • Igor P. Maurell FURG
  • Caue Ferreira ITA
  • Carlos A. Eguti ITA
  • Paulo Drews-Jr FURG

Resumo


Geometric primitive fitting in point clouds of large structures is an important task in many scenarios, such as: 3D data simplification, scene reconstruction and 3D CAD reverse engineering. With the advance in unmanned aerial vehicle technologies, multiple large-scale 3D point clouds are being generated by these systems for many goals. Some of the solutions that emerged with this 3D data high availability can be directly or indirectly benefited by a pipeline that finds the geometric primitives underlying in large-scale 3D point clouds. Thus, the work present a pipeline to completely process those data and fit simple geometric primitives to it. Results showed the capabilities of this type of technique in real world data.
Palavras-chave: Measurement, Three-dimensional displays, Fitting, Pipelines, Reverse engineering, Education, Random access memory
Publicado
11/10/2021
MAURELL, Igor P.; FERREIRA, Caue; EGUTI, Carlos A.; DREWS-JR, Paulo. Geometric Primitive Fitting in Large Structure 3D Point Clouds Acquired by Drones. In: SIMPÓSIO BRASILEIRO DE ROBÓTICA E SIMPÓSIO LATINO AMERICANO DE ROBÓTICA (SBR/LARS), 13. , 2021, Online. Anais [...]. Porto Alegre: Sociedade Brasileira de Computação, 2021 . p. 108-113.