Détail de l'autorité
1-Pas de projet /
Nom :
1-Pas de projet
titre complet :
sans projet
|
Documents disponibles (240)
Ajouter le résultat dans votre panier
Visionner les documents numériques
Affiner la recherche Interroger des sources externes
Titre : Point cloud oversegmentation with graph-structured deep metric learning Type de document : Article/Communication Auteurs : Loïc Landrieu , Auteur ; Mohamed Boussaha , Auteur Editeur : Computer vision foundation CVF Année de publication : 2019 Projets : 1-Pas de projet / Conférence : CVPR 2019, IEEE Conference on Computer Vision and Pattern Recognition 16/06/2019 20/06/2019 Long Beach Californie - Etats-Unis Open Access Proceedings Importance : pp 7432 - 7441 Format : 21 x 30 cm Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image
[Termes IGN] apprentissage profond
[Termes IGN] données localisées 3D
[Termes IGN] graphe
[Termes IGN] réseau neuronal artificiel
[Termes IGN] segmentation sémantique
[Termes IGN] semis de pointsRésumé : (auteur) We propose a new supervized learning framework foroversegmenting 3D point clouds into superpoints. We castthis problem as learning deep embeddings of the local ge-ometry and radiometry of 3D points, such that the border ofobjects presents high contrasts. The embeddings are com-puted using a lightweight neural network operating on thepoints’ local neighborhood. Finally, we formulate pointcloud oversegmentation as a graph partition problem withrespect to the learned embeddings.This new approach allows us to set a new state-of-the-artin point cloud oversegmentation by a significant margin, ona dense indoor dataset (S3DIS) and a sparse outdoor one(vKITTI). Our best solution requires over five times fewersuperpoints to reach similar performance than previouslypublished methods on S3DIS. Furthermore, we show thatour framework can be used to improve superpoint-basedsemantic segmentation algorithms, setting a new state-of-the-art for this task as well. Numéro de notice : C2019-017 Affiliation des auteurs : LASTIG MATIS (2012-2019) Autre URL associée : vers CVF Thématique : IMAGERIE/INFORMATIQUE Nature : Communication nature-HAL : ComAvecCL&ActesPubliésIntl DOI : 10.1109/CVPR.2019.00762 Date de publication en ligne : 09/01/2020 En ligne : https://doi.org/10.1109/CVPR.2019.00762 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=93351 Reconciling upper mantle seismic velocity and density structure below ocean basins / Isabelle Panet (2019)
Titre : Reconciling upper mantle seismic velocity and density structure below ocean basins Type de document : Article/Communication Auteurs : Isabelle Panet , Auteur ; Barbara Romanowicz, Auteur ; Marianne Greff-Lefftz, Auteur Editeur : Saint-Mandé : Institut national de l'information géographique et forestière - IGN (2012-) Année de publication : 2019 Projets : 1-Pas de projet / Conférence : AGU 2019 Fall Meeting 09/12/2019 13/12/2019 San Francisco Californie - Etats-Unis programme sans actes Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Géodésie physique
[Termes IGN] champ de pesanteur terrestre
[Termes IGN] déformation de la croute terrestre
[Termes IGN] données bathymétriques
[Termes IGN] données GRACE
[Termes IGN] fond marin
[Termes IGN] geoïde marin
[Termes IGN] géophysique interne
[Termes IGN] Indien (océan)
[Termes IGN] manteau terrestre
[Termes IGN] Pacifique (océan)
[Termes IGN] structure géologique
[Termes IGN] vitesse de déplacementRésumé : (auteur) Imaging the spatial pattern of mantle flows and constraining their mass is one of the keys to understand the character of mantle convection inside the Earth, and its interactions with plate motions. The horizontal planform of the flows, their heterogeneity and mass transport at depth, are reflected in variations of the gravity field and seismic velocities, as well as deformations of the Earth's surface. Over ocean basins, these observables show an elusive medium-scale structure. A 1500-2000 km wavelength directional fabric following the present-day absolute plate motion is present in the Pacific Ocean in GRACE satellite gravity data (Hayn et al., 2012), while 2000-km wavelength slow shear velocity anomalies sharing a similar orientation are found in seismic tomography at upper mantle depths below the oceans (SEMUM2, French et al., 2013). Today, the dynamic processes at the origin of these observations remain unresolved.
Here, we develop a joint analysis of satellite gravity and bathymetry data together with the SEMUM2 seismic tomography model, in order to advance our understanding of upper to mid-mantle flows below the oceans. First, we enhance and reconstruct the medium-scale gravity and seafloor topography signals aligned with the present-day plate motion from an analysis of the rates of gravity vector variations and seafloor slopes. Then, we compare the obtained signals with the spatial distribution of shear velocity anomalies at depth. We show that slow velocity anomalies coincide with geoid lows, depressions in the seafloor topography, and mass excess in the mantle, in the Pacific ocean and part of the Indian ocean. We first consider a purely thermal interpretation of the seismic velocity variations, associated with medium-scale convective rolls in the upper to mid-mantle, a process able to only explain the observed geometry of anomalies. Investigating whether the needed mass excess arises from lithospheric or deeper sources, such as at the level of the 660-km interface, we conclude that it lies more likely within the slow velocity anomalies themselves, suggesting hot and dense structures. We finally discuss the possible meaning and implications of these results.Numéro de notice : C2019-058 Affiliation des auteurs : Géodésie+Ext (mi2018-2019) Thématique : POSITIONNEMENT Nature : Poster nature-HAL : Poster-avec-CL DOI : sans Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=96896
Titre : Scalable evaluation of 3D city models Type de document : Article/Communication Auteurs : Oussama Ennafii , Auteur ; Arnaud Le Bris , Auteur ; Florent Lafarge, Auteur ; Clément Mallet , Auteur Editeur : Saint-Mandé : Institut national de l'information géographique et forestière - IGN (2012-) Année de publication : 2019 Projets : 1-Pas de projet / Conférence : IGARSS 2019, IEEE International Geoscience And Remote Sensing Symposium 28/07/2019 02/08/2019 Yokohama Japon Proceedings IEEE Importance : 4 p. Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image
[Termes IGN] analyse de données
[Termes IGN] classification dirigée
[Termes IGN] fusion d'images
[Termes IGN] fusion de données
[Termes IGN] image à très haute résolution
[Termes IGN] modélisation 3D du bâti BIM
[Termes IGN] taxinomieRésumé : (Auteur) The generation of 3D building models from Very High Resolution geospatial data is now an automatized procedure. However, urban areas are very complex and practitioners still have to visually assess the correctness of these models and detect reconstruction errors. We proposed an approach for automatically evaluating the quality of 3D building models. It is cast as a supervised classification task based on a hierarchical taxonomy and multimodal handcrafted features (building geometry, optical images, height data). In this paper, we evaluate how the urban area composition impacts prediction transferability and scalability of our framework to unseen scenes. This allows to define minimal feature and training sets for a problem where no benchmark data has been released so far. Numéro de notice : C2019-006 Affiliation des auteurs : LASTIG MATIS+Ext (2012-2019) Autre URL associée : vers HAL Thématique : IMAGERIE/URBANISME Nature : Poster nature-HAL : Poster-avec-CL DOI : 10.1109/IGARSS.2019.8899337 Date de publication en ligne : 14/11/2019 En ligne : https://doi.org/10.1109/IGARSS.2019.8899337 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=92592 SUMAC 2019, 1st workshop on Structuring and Understanding of Multimedia heritAge Contents / Valérie Gouet-Brunet (2019)
Titre : SUMAC 2019, 1st workshop on Structuring and Understanding of Multimedia heritAge Contents Type de document : Article/Communication Auteurs : Valérie Gouet-Brunet , Éditeur scientifique ; Margarita Khokhlova , Éditeur scientifique ; Liming Chen, Éditeur scientifique ; Sander Münster, Éditeur scientifique ; Sander Münster Editeur : New York [Etats-Unis] : Association for computing machinery ACM Année de publication : 2019 Projets : 1-Pas de projet / Conférence : MM 2019, 27th ACM International Conference on Multimedia 21/10/2019 25/10/2019 Nice France Proceedings ACM Importance : pp 2726 - 2727 Format : 21 x 30 cm Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Systèmes d'information
[Termes IGN] bibliothèque numérique
[Termes IGN] données massives
[Termes IGN] interface homme-machine
[Termes IGN] multimedia
[Termes IGN] patrimoine culturel
[Termes IGN] patrimoine immobilier
[Termes IGN] recherche scientifique
[Termes IGN] valorisationRésumé : (auteur) SUMAC 2019 is the first workshop on Structuring and Understanding of Multimedia heritAge Contents. It is held in Nice, France on October 21, 2019 and is co-located with the 27th ACM International Conference on Multimedia. Its objective is to present and discuss the latest and most significant trends and challenges in the analysis, structuring and understanding of multimedia contents dedicated to the valorization of heritage, with the emphasis on the unlocking of and access to the big data of the past. A representative scope of Computer Science methodologies dedicated to the processing of multimedia heritage contents and their exploitation is covered by the works presented, with the ambition of advancing and raising awareness about this fully developing research field. Numéro de notice : 25324 Affiliation des auteurs : LASTIG MATIS+Ext (2012-2019) Thématique : GEOMATIQUE/IMAGERIE Nature : Communication nature-HAL : ComAvecCL&ActesPubliésIntl DOI : 10.1145/3343031.3350554 En ligne : https://doi.org/10.1145/3343031.3350554 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=93622
Titre : Supervized segmentation with graph-structured deep metric learning Type de document : Article/Communication Auteurs : Loïc Landrieu , Auteur ; Mohamed Boussaha , Auteur Editeur : Ithaca [New York - Etats-Unis] : ArXiv - Université Cornell Année de publication : 2019 Projets : 1-Pas de projet / Conférence : ICML 2019, Workshop on Learning and Reasoning with Graph-Structured Representations in International Conference on Machine Learning 15/06/2019 15/06/2019 Long Beach Californie - Etats-Unis Open Access Proceedings Importance : 15 p. Langues : Français (fre) Descripteur : [Vedettes matières IGN] Intelligence artificielle
[Termes IGN] apprentissage profond
[Termes IGN] graphe
[Termes IGN] segmentation
[Termes IGN] semis de pointsRésumé : (auteur) We present a fully-supervized method for learning to segment data structured by an adjacency graph. We introduce the graph-structured contrastive loss, a loss function structured by a ground truth segmentation. It promotes learning vertex embeddings which are homogeneous within desired segments, and have high contrast at their interface. Thus, computing a piecewise-constant approximation of such embeddings produces a graph-partition close to the objective segmentation. This loss is fully backpropagable, which allows us to learn vertex embeddings with deep learning algorithms. We evaluate our methods on a 3D point cloud oversegmentation task, defining a new state-of-the-art by a large margin. These results are based on the published work of Landrieu and Boussaha 2019. Numéro de notice : C2019-050 Affiliation des auteurs : LASTIG MATIS (2012-2019) Autre URL associée : vers ArXiv Nature : Poster nature-HAL : Poster-avec-CL DOI : 10.48550/arXiv.1905.04014 Date de publication en ligne : 19/05/2019 En ligne : https://graphreason.github.io/papers/4.pdf Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=92819 The necessary yet complex evaluation of 3D city models: a semantic approach / Oussama Ennafii (2019)PermalinkPermalinkTime-space tradeoff in deep learning models for crop classification on satellite multi-spectral image time series / Vivien Sainte Fare Garnot (2019)PermalinkTowards improving knowledge capitalization system for sport events legacy / Malika Grim-Yefsah (2019)PermalinkLe vandalisme dans l’information géographique volontaire, détection de l’IG volontaire vandalisée : du concept à la détection non supervisée d’anomalie / Quy Thy Truong in Revue internationale de géomatique, vol 29 n° 1 (janvier - mars 2019)PermalinkLa forme de la terre dans l'histoire occidentale / Xavier Della Chiesa in XYZ, n° 157 (décembre 2018 - février 2019)PermalinkThe reviewing process for ISPRS events / Clément Mallet in ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, vol IV-5 (November 2018)PermalinkLa filiera foresta-legno francese tra potenziale di mitigazione dei cambiamenti climatici e necessità di adattamento / Philippe Delacote in Agriregionieuropa, anno 14 n° 54 (2018)PermalinkPermalinkUnmixing polarimetric radar images based on land cover type identified by higher resolution optical data before target decomposition: application to forest and bare soil / Sébastien Giordano in IEEE Transactions on geoscience and remote sensing, vol 56 n° 10 (October 2018)Permalink3D urban geovisualization: in situ augmented and mixed reality experiments / Alexandre Devaux in ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, vol IV-4 (October 2018)PermalinkGNSS-assisted integrated sensor orientation with sensor pre-calibration for accurate corridor mapping / Yilin Zhou in Sensors, vol 18 n° 9 (September 2018)PermalinkA quelles altitudes se trouvent les horloges atomiques de l'observatoire de Paris ? / Xavier Collilieux in XYZ, n° 156 (septembre - novembre 2018)PermalinkEst-il possible de tirer des enseignements des introductions anciennes d'agents pathogènes ? L'exemple de la graphiose de l'orme / Dominique Piou in Revue forestière française, vol 70 n° 6 (2018)PermalinkForeword to the special issue on urban remote sensing for smarter cities / Prashanth Reddy Marpu in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol 11 n° 8 (August 2018)PermalinkUnsupervised detection of ruptures in spatial relationships in video sequences based on log‑likelihood ratio / Abdalbassir Abou-Elailah in Pattern Analysis and Applications, vol 21 n° 3 (August 2018)PermalinkFrom hierarchy to networking: the evolution of the “twenty-first-century Maritime Silk Road” container shipping system / Liehui Wang in Transport reviews, vol 38 n° 4 ([01/07/2018])PermalinkSecond iteration of photogrammetric processing to refine image orientation with improved tie-points / Truong Giang Nguyen in Sensors, vol 18 n° 7 (July 2018)PermalinkSoil moisture estimation in Ferlo region (Senegal) using radar (ENVISAT/ASAR) and optical (SPOT/VEGETATION) data / Gayane Faye in The Egyptian Journal of Remote Sensing and Space Science, Vol. 21 suppl.1 (juillet 2018)PermalinkClassification à très large échelle d’images satellites à très haute résolution spatiale par réseaux de neurones convolutifs / Tristan Postadjian in Revue Française de Photogrammétrie et de Télédétection, n° 217-218 (juin - septembre 2018)Permalink