Weakly supervised segmentation-aided classification of urban scenes from 3D LiDAR point clouds

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 anthropic 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

Cited literature [19 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01497548
Contributor : Loic Landrieu <>
Submitted on : Tuesday, May 9, 2017 - 10:19:20 AM
Last modification on : Friday, May 12, 2017 - 10:17:22 AM
Document(s) archivé(s) le : Thursday, August 10, 2017 - 12:31:55 PM

File

ISPRS2017_2.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01497548, version 3

Citation

Stéphane Guinard, Loic Landrieu. Weakly supervised segmentation-aided classification of urban scenes from 3D LiDAR point clouds. ISPRS Workshop 2017, Jun 2017, Hannover, Germany. ⟨hal-01497548v3⟩

Share

Metrics

Record views

281

Files downloads

397