Centre de documentation
scientifique

MS-RRFSegNetMultiscale regional relation feature segmentation network for semantic segmentation of urban scene point clouds / Haifeng Luo in IEEE Transactions on geoscience and remote sensing, Vol 58 n° 12 (December 2020)  

Public [article]  inIEEE Transactions on geoscience and remote sensing > Vol 58 n° 12 (December 2020) . - pp 8301 - 8315 Titre : MS-RRFSegNetMultiscale regional relation feature segmentation network for semantic segmentation of urban scene point clouds Type de document : Article/Communication Auteurs : Haifeng Luo, Auteur ; Chongcheng Chen, Auteur ; Lina Fang, Auteur ; et al., Auteur Année de publication : 2020 Article en page(s) : pp 8301 - 8315 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Lasergrammétrie [Termes IGN] apprentissage profond [Termes IGN] cognition [Termes IGN] données lidar [Termes IGN] extraction de traits caractéristiques [Termes IGN] représentation multiple [Termes IGN] scène urbaine [Termes IGN] segmentation sémantique [Termes IGN] semis de points Résumé : (auteur) Semantic segmentation is one of the fundamental tasks in understanding and applying urban scene point clouds. Recently, deep learning has been introduced to the field of point cloud processing. However, compared to images that are characterized by their regular data structure, a point cloud is a set of unordered points, which makes semantic segmentation a challenge. Consequently, the existing deep learning methods for semantic segmentation of point cloud achieve less success than those applied to images. In this article, we propose a novel method for urban scene point cloud semantic segmentation using deep learning. First, we use homogeneous supervoxels to reorganize raw point clouds to effectively reduce the computational complexity and improve the nonuniform distribution. Then, we use supervoxels as basic processing units, which can further expand receptive fields to obtain more descriptive contexts. Next, a sparse autoencoder (SAE) is presented for feature embedding representations of the supervoxels. Subsequently, we propose a regional relation feature reasoning module (RRFRM) inspired by relation reasoning network and design a multiscale regional relation feature segmentation network (MS-RRFSegNet) based on the RRFRM to semantically label supervoxels. Finally, the supervoxel-level inferences are transformed into point-level fine-grained predictions. The proposed framework is evaluated in two open benchmarks (Paris-Lille-3D and Semantic3D). The evaluation results show that the proposed method achieves competitive overall performance and outperforms other related approaches in several object categories. An implementation of our method is available at: https://github.com/HiphonL/MS_RRFSegNet . Numéro de notice : A2020-738 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2020.2985695 Date de publication en ligne : 28/04/2020 En ligne : https://doi.org/10.1109/TGRS.2020.2985695 Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=96363 [article]