A deep perceptual metric for 3D point clouds - Université d'Orléans Accéder directement au contenu
Communication Dans Un Congrès Année : 2021

A deep perceptual metric for 3D point clouds

A deep perceptual metric for 3D point clouds

Résumé

Point clouds are essential for storage and transmission of 3D content. As they can entail significant volumes of data, point cloud compression is crucial for practical usage. Recently, point cloud geometry compression approaches based on deep neural networks have been explored. In this paper, we evaluate the ability to predict perceptual quality of typical voxel-based loss functions employed to train these networks. We find that the commonly used focal loss and weighted binary cross entropy are poorly correlated with human perception. We thus propose a perceptual loss function for 3D point clouds which outperforms existing loss functions on the ICIP2020 subjective dataset. In addition, we propose a novel truncated distance field voxel grid representation and find that it leads to sparser latent spaces and loss functions that are more correlated with perceived visual quality compared to a binary representation. The source code is available at https: // github. com/ mauriceqch/ 2021_ pc_ perceptual_ loss .
Fichier principal
Vignette du fichier
EI_2021___A_deep_perceptual_metric_for_3D_point_clouds.pdf (394.44 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03108242 , version 1 (06-02-2021)

Identifiants

Citer

Maurice Quach, Aladine Chetouani, Giuseppe Valenzise, Frédéric Dufaux. A deep perceptual metric for 3D point clouds. Image Quality and System Performance, IS&T International Symposium on Electronic Imaging (EI 2021), Jan 2021, San Francisco, United States. pp.257-1-257-7, ⟨10.2352/ISSN.2470-1173.2021.9.IQSP-257⟩. ⟨hal-03108242⟩
108 Consultations
302 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More