Las Vegas Surface Reconstruction Toolkit 2.0 (LVR2)

The Las Vegas Surface Reconstruction Toolkit is an Open Source toolkit to reconstruct triangle meshes from unordered point clouds. It supports the generation of textured models either using colored point clouds or point clouds plus images and according calibration matrices. The LVR toolkit provides an Open Source C++ API for meshing and texture generation as well as an I/O interface to store the generated meshes in different exchange formats including Stanford PLY, Wavefront OBJ and Collada. In contrast to other meshing software, it focuses on reconstruction of large scale environments on city scale from high resolution point clouds.

Besides meshing algorithms LVR provides algorithms to post-process the generated meshes in the context of Mobile Robotics, especially Semantic Mapping. This includes algorithms for mesh decimation, segmentation and cluster labeling. The included viewer application can be used to visualize the reconstruction results.

To use the reconstructions on mobile robots, we provide an ROS interface to serialize the reconstruction and Plugins for RViz to display the data and interactive segmentation ROS Mesh Tools

Features

Welcome to the features of LVR2:

  • Registration
  • Normal Estimation
  • Reconstruction
  • Mesh Optimization
  • Texturizing

Other useful features:

  • Raycasting
  • Large Scale Reconstruction
  • Visualization

Recent Publications

Quickly discover relevant content by filtering publications.

In this paper, we present a reference data set that maps hyperspectral intensity data to a terrestrial 3D laser scanner to generate …

This thesis presents a method to automatically produce compressed polygonal meshes of arbitrary environments using a modifed Marching …

In this paper we present an extension of Large Scale Kinect Fusion to compute optimized triangle meshes on-the-fly by removing …

Projects

*

3DinOS

In quite a few projects world-wide, 3D Mapping technology has been used for documenting historical buildings – typically …