3D Scene Modeling from Dense Video Light Fields

Xiaoran Jiang (1) , Christian Galea (1) , Laurent Guillo (2, 1) , Christine Guillemot (1)

1 Sirocco - Analysis representation, compression and communication of visual data
2 CNRS - Centre National de la Recherche Scientifique

Light field imaging offers unprecedented opportunities for advanced scene analysis and modelling, with potential applications in various domains such as augmented reality, 3D robotics, and microscopy. This paper illustrates the potential of dense video light fields for 3D scene modeling. We first recall the principles of plenoptic cameras and present a downloadable test dataset captured with a Raytrix 2.0 plenop-tic camera. Methods to estimate the scene depth and to construct a 3D point cloud representation of the scene from the captured light field are then described. I

Mots clés: Light field video; plenoptic cameras; depth maps; 3D point clouds.

Domaines: Informatique [cs]

Xiaoran Jiang, Christian Galea, Laurent Guillo, Christine Guillemot. 3D Scene Modeling from Dense Video Light Fields. IC3D 2018 - 8th International conference on 3D immersion, Dec 2018, Brussels, Belgium. pp.1-7. ⟨hal-01925319⟩