Centre de documentation
scientifique

Improving deep learning on point cloud by maximizing mutual information across layers / Di Wang in Pattern recognition, vol 131 (November 2022)  

Public [article]  inPattern recognition > vol 131 (November 2022) . - n° 108892 Titre : Improving deep learning on point cloud by maximizing mutual information across layers Type de document : Article/Communication Auteurs : Di Wang, Auteur ; Lulu Tang, Auteur ; Xu Wang, Auteur ; et al., Auteur Année de publication : 2022 Article en page(s) : n° 108892 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] apprentissage profond [Termes IGN] classification par réseau neuronal convolutif [Termes IGN] détection d'objet [Termes IGN] entropie de Shannon [Termes IGN] information sémantique [Termes IGN] segmentation sémantique [Termes IGN] semis de points [Termes IGN] transformation géométrique [Termes IGN] vision par ordinateur [Termes IGN] visualisation 3D Résumé : (auteur) It is a fundamental and vital task to enhance the perception capability of the point cloud learning network in 3D machine vision applications. Most existing methods utilize feature fusion and geometric transformation to improve point cloud learning without paying enough attention to mining further intrinsic information across multiple network layers. Motivated to improve consistency between hierarchical features and strengthen the perception capability of the point cloud network, we propose exploring whether maximizing the mutual information (MI) across shallow and deep layers is beneficial to improve representation learning on point clouds. A novel design of Maximizing Mutual Information (MMI) Module is proposed, which assists the training process of the main network to capture discriminative features of the input point clouds. Specifically, the MMI-based loss function is employed to constrain the differences of semantic information in two hierarchical features extracted from the shallow and deep layers of the network. Extensive experiments show that our method is generally applicable to point cloud tasks, including classification, shape retrieval, indoor scene segmentation, 3D object detection, and completion, and illustrate the efficacy of our proposed method and its advantages over existing ones. Our source code is available at https://github.com/wendydidi/MMI.git. Numéro de notice : A2022-780 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article DOI : https://doi.org/10.1016/j.patcog.2022.108892 Date de publication en ligne : 08/07/2022 En ligne : https://doi.org/10.1016/j.patcog.2022.108892 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=101859 [article]