Pré-segmentation pour la classification faiblement supervisée de scènes urbaines à partir de nuages de points 3D LIDAR

Abstract : We consider the problem of the semantic classification of 3D LiDAR point clouds obtained from urban scenes when the training set is limited. We propose a non-parametric segmentation model for urban scenes composed of anthro-pic objects of simple shapes, partionning the scene into geometrically-homogeneous segments which size is determined by the local complexity. This segmentation can be integrated into a conditional random field classifier (CRF) in order to capture the high-level structure of the scene. For each cluster, this allows us to aggregate the noisy predictions of a weakly-supervised classifier to produce a higher confidence data term. We demonstrate the improvement provided by our method over two publicly-available large-scale data sets.
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-01499571
Contributor : Loic Landrieu <>
Submitted on : Monday, April 3, 2017 - 5:56:39 PM
Last modification on : Friday, April 14, 2017 - 9:39:55 AM
Document(s) archivé(s) le : Tuesday, July 4, 2017 - 12:27:03 PM

File

ORASIS_3.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01499571, version 1

Citation

Stéphane Guinard, Loic Landrieu, Bruno Vallet. Pré-segmentation pour la classification faiblement supervisée de scènes urbaines à partir de nuages de points 3D LIDAR. 2017. ⟨hal-01499571v1⟩

Share

Metrics

Record views

43

Files downloads

46