Technical Report: Co-learning of geometry and semantics for online 3D mapping - Archive ouverte HAL Accéder directement au contenu
Rapport (Rapport De Recherche) Année : 2019

Technical Report: Co-learning of geometry and semantics for online 3D mapping

Résumé

This paper is a technical report about our submission for the ECCV 2018 3DRMS Workshop Challenge on Semantic 3D Reconstruction \cite{Tylecek2018rms}. In this paper, we address 3D semantic reconstruction for autonomous navigation using co-learning of depth map and semantic segmentation. The core of our pipeline is a deep multi-task neural network which tightly refines depth and also produces accurate semantic segmentation maps. Its inputs are an image and a raw depth map produced from a pair of images by standard stereo vision. The resulting semantic 3D point clouds are then merged in order to create a consistent 3D mesh, in turn used to produce dense semantic 3D reconstruction maps. The performances of each step of the proposed method are evaluated on the dataset and multiple tasks of the 3DRMS Challenge, and repeatedly surpass state-of-the-art approaches.
Fichier principal
Vignette du fichier
eccv2018submission.pdf (7.73 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-02341718 , version 1 (31-10-2019)

Identifiants

Citer

Marcela Carvalho, Maxime Ferrera, Alexandre Boulch, Julien Moras, Bertrand Le Saux, et al.. Technical Report: Co-learning of geometry and semantics for online 3D mapping. [Research Report] ONERA. 2019. ⟨hal-02341718⟩
43 Consultations
25 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More