Real-time 3D modeling with a RGB-D camera and on-board processing


Abstract:

In this article we present a three dimensional modeling system that generates precise real-time mapping using a RGB-D camera. With the use of the light weight sensors Microsoft Kinect and small and powerful computers like the Intel Stick Core M3 Processor, our system can run all the computation and sensing required to smoothly run SLAM (Simultaneous Localization and Mapping) on-board and in real-time, removing the dependence on unreliable wireless communication. We use visual odometry, loop closure and graph optimization. Our approach is capable of generating accurate maps of several objects analyzing the data yielded by several tests of the system.

Año de publicación:

2017

Keywords:

  • Loop closure detection
  • UAVs
  • SLAM
  • Graph optimization
  • Visual odometry
  • RANSAC
  • RGB-D

Fuente:

googlegoogle
scopusscopus

Tipo de documento:

Conference Object

Estado:

Acceso restringido

Áreas de conocimiento:

  • Ciencias de la computación
  • Visión por computadora

Áreas temáticas:

  • Métodos informáticos especiales
  • Física aplicada