English

Surface Reconstruction from 3D Line Segments

Computer Vision and Pattern Recognition 2019-11-04 v1

Abstract

In man-made environments such as indoor scenes, when point-based 3D reconstruction fails due to the lack of texture, lines can still be detected and used to support surfaces. We present a novel method for watertight piecewise-planar surface reconstruction from 3D line segments with visibility information. First, planes are extracted by a novel RANSAC approach for line segments that allows multiple shape support. Then, each 3D cell of a plane arrangement is labeled full or empty based on line attachment to planes, visibility and regularization. Experiments show the robustness to sparse input data, noise and outliers.

Keywords

Cite

@article{arxiv.1911.00451,
  title  = {Surface Reconstruction from 3D Line Segments},
  author = {Pierre-Alain Langlois and Alexandre Boulch and Renaud Marlet},
  journal= {arXiv preprint arXiv:1911.00451},
  year   = {2019}
}

Comments

In 3DV 2019 (Oral)

R2 v1 2026-06-23T12:02:25.050Z