Publications du LaSTIG
Les publications antérieures au LaSTIG sont celles des laboratoires qui ont formé le LaSTIG : COGIT, LOEMI et MATIS, à l'exception du LAREG - Vous pouvez affiner la recherche au sein des références
![](./images/expand_all.gif)
![](./images/collapse_all.gif)
![Imprimer...](./images/print.gif)
PSMNet-FusionX3 : LiDAR-guided deep learning stereo dense matching on aerial images / Teng Wu (2023)
![]()
Titre : PSMNet-FusionX3 : LiDAR-guided deep learning stereo dense matching on aerial images Type de document : Article/Communication Auteurs : Teng Wu , Auteur ; Bruno Vallet
, Auteur ; Marc Pierrot-Deseilligny
, Auteur
Editeur : Computer vision foundation CVF Année de publication : 2023 Conférence : CVPR 2023, IEEE Conference on Computer Vision and Pattern Recognition workshops 18/06/2023 22/06/2023 Vancouver Colombie britannique - Canada OA Proceedings Importance : pp 6526 - 6535 Note générale : bibliographie
voir aussi https://openaccess.thecvf.com/content/CVPR2023W/PCV/supplemental/Wu_PSMNet-FusionX3_LiDAR-Guided_Deep_CVPRW_2023_supplemental.pdfLangues : Anglais (eng) Descripteur : [Vedettes matières IGN] Applications photogrammétriques
[Termes IGN] appariement dense
[Termes IGN] apprentissage profond
[Termes IGN] chaîne de traitement
[Termes IGN] données lidar
[Termes IGN] données localisées 3D
[Termes IGN] image aérienne à axe vertical
[Termes IGN] scène 3D
[Termes IGN] Triangulated Irregular NetworkRésumé : (auteur) Dense image matching (DIM) and LiDAR are two complementary techniques for recovering the 3D geometry of real scenes. While DIM provides dense surfaces, they are often noisy and contaminated with outliers. Conversely, LiDAR is more accurate and robust, but less dense and more expensive compared to DIM. In this work, we investigate learning-based methods to refine surfaces produced by photogrammetry with sparse LiDAR point clouds. Unlike the current state-of-the-art approaches in the computer vision community, our focus is on aerial acquisitions typical in photogrammetry. We propose a densification pipeline that adopts a PSMNet backbone with triangulated irregular network interpolation based expansion, feature enhancement in cost volume, and conditional cost volume normalization, i.e. PSMNet-FusionX3. Our method works better on low density and is less sensitive to distribution, demonstrating its effectiveness across a range of LiDAR point cloud densities and distributions, including analyses of dataset shifts. Furthermore, we have made both our aerial (image and disparity) dataset and code available for public use. Further information can be found at https://github.com/ whuwuteng/PSMNet-FusionX3. Numéro de notice : C2023-006 Affiliation des auteurs : UGE-LASTIG (2020- ) Thématique : IMAGERIE/INFORMATIQUE Nature : Communication DOI : sans En ligne : https://openaccess.thecvf.com/content/CVPR2023W/PCV/papers/Wu_PSMNet-FusionX3_Li [...] Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=103277 SALT: A multifeature ensemble learning framework for mapping urban functional zones from VGI data and VHR images / Hao Wu in Computers, Environment and Urban Systems, vol 100 (March 2023)
![]()
[article]
Titre : SALT: A multifeature ensemble learning framework for mapping urban functional zones from VGI data and VHR images Type de document : Article/Communication Auteurs : Hao Wu, Auteur ; Wenting Luo, Auteur ; Anqi Lin, Auteur ; Fanghua Hao, Auteur ; Ana-Maria Olteanu-Raimond , Auteur ; Lanfa Liu, Auteur ; Yan Li, Auteur
Année de publication : 2023 Projets : 1-Pas de projet / Article en page(s) : n° 101921 Note générale : Bibliographie
This work was supported by the National Natural Science Foundation of China [42201468, 42071358], Postdoctoral Innovation Talents Support Program of China [BX20220128], China Postdoctoral Science Foundation [2022M721283] and Fundamental Research Funds for the Central Universities [CCNU22QN018].Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Analyse spatiale
[Termes IGN] analyse multicritère
[Termes IGN] apprentissage automatique
[Termes IGN] boosting adapté
[Termes IGN] cartographie urbaine
[Termes IGN] Chine
[Termes IGN] détection du bâti
[Termes IGN] données localisées des bénévoles
[Termes IGN] image à très haute résolution
[Termes IGN] morphologie urbaine
[Termes IGN] OpenStreetMap
[Termes IGN] point d'intérêt
[Termes IGN] représentation spatiale
[Termes IGN] zone urbaineRésumé : (auteur) Urban functional zone mapping is essential for providing deeper insights into urban morphology and improving urban planning. The emergence of Volunteered Geographic Information (VGI), which provides abundant semantic data, offers a great opportunity to enrich land use information extracted from remote sensing (RS) images. Taking advantage of very-high-resolution (VHR) images and VGI data, this work proposed a SATL multifeature ensemble learning framework for mapping urban functional zones that integrated 65 features from the shapes of building objects, attributes of points of interest (POIs) tags, locations of cellphone users and textures of VHR images. The dimensionality of SALT features was reduced by the autoencoder, and the compressed features were applied to train the ensemble learning model composed of multiple classifiers for optimizing the urban functional zone classification. The effectiveness of the proposed framework was tested in an urbanized region of Nanchang City. The results indicated that the SALT features considering population dynamics and building shapes are comprehensive and feasible for urban functional zone mapping. The autoencoder has been proven efficient for dimension reduction of the original SALT features as it significantly improves the classification of urban functional zones. Moreover, the ensemble learning outperforms other machine learning models in terms of the accuracy and robustness when dealing with multi-classification tasks. Numéro de notice : A2023-125 Affiliation des auteurs : UGE-LASTIG+Ext (2020- ) Thématique : GEOMATIQUE/IMAGERIE/INFORMATIQUE Nature : Article DOI : 10.1016/j.compenvurbsys.2022.101921 Date de publication en ligne : 06/12/2022 En ligne : https://doi.org/10.1016/j.compenvurbsys.2022.101921 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=102504
in Computers, Environment and Urban Systems > vol 100 (March 2023) . - n° 101921[article]Semi-automated Pipeline to Produce Customizable Tactile Maps of Street Intersections for People with Visual Impairments / Yuhao Jiang (2023)
![]()
Titre : Semi-automated Pipeline to Produce Customizable Tactile Maps of Street Intersections for People with Visual Impairments Type de document : Article/Communication Auteurs : Yuhao Jiang, Auteur ; María-Jesús Lobo , Auteur ; Sidonie Christophe
, Auteur ; Christophe Jouffrais, Auteur
Editeur : Göttingen : Copernicus publications Année de publication : 2023 Collection : AGILE GIScience Series num. 4 Conférence : AGILE 2023, 26th international AGILE Conference on Geographic Information Science, Spatial data for design 13/06/2023 16/06/2023 Delft Pays-Bas OA Proceedings Importance : n° 29 ; 8 p. Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Cartographie
[Termes IGN] carrefour
[Termes IGN] carte sur mesure
[Termes IGN] carte tactile
[Termes IGN] chaîne de traitement
[Termes IGN] OpenStreetMap
[Termes IGN] personne malvoyanteIndex. décimale : 39.00 Cartographie - généralités - Cartologie Résumé : (auteur) Street intersections are very challenging for people with visual impairments. Manually produced tactile maps are an important support in teaching and assisting independent journeys as they can be customized to serve the visually impaired audience with diverse tactile reading and mobility skills in different use scenarios. But the manual map production involves a huge workload that makes the maps less accessible. This paper explores the possibility of semi-automatically producing customizable tactile maps for street intersections. It presents a parameterized semi-automated pipeline based on OSM data that allows the maps to be customized in size, map features, geometry processing choices, and symbolizations. It produces street intersection maps in two scales of three sizes, with different levels of details and styles. Numéro de notice : C2023-013 Affiliation des auteurs : UGE-LASTIG+Ext (2020- ) Thématique : GEOMATIQUE/INFORMATIQUE Nature : Communication nature-HAL : ComAvecCL&ActesPubliésIntl DOI : 10.5194/agile-giss-4-29-2023 Date de publication en ligne : 06/06/2023 En ligne : https://doi.org/10.5194/agile-giss-4-29-2023 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=103307
Titre : Structured learning of geospatial data Type de document : Thèse/HDR Auteurs : Loïc Landrieu , Auteur
Editeur : Champs-sur-Marne [France] : Université Gustave Eiffel Année de publication : 2023 Importance : 179 p. Format : 21 x 30 cm Note générale : Bibliographie
Habilitation à Diriger des Recherches délivrée par l'Université Gustave Eiffel, Spécialité "Sciences et Technologies de l'Information Géographique"Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique
[Termes IGN] algorithme Cut Pursuit
[Termes IGN] apprentissage automatique
[Termes IGN] carte agricole
[Termes IGN] graphe
[Termes IGN] lasergrammétrie
[Termes IGN] reconnaissance de formes
[Termes IGN] segmentation sémantique
[Termes IGN] série temporelle
[Termes IGN] vision par ordinateurRésumé : (auteur) This manuscript presents an overview of my work in the field of geospatial machine learning, a rapidly growing interdisciplinary field that poses many methodological challenges and has a wide range of impactful applications. Throughout my research, I have focused on developing bespoke approaches that leverage the unique properties of geospatial data to create more efficient, precise, and parsimonious models. This manuscript is divided into four main chapters, each covering a different property of geospatial data structures that can be leveraged algorithmically. The first chapter presents a versatile mathematical framework formalizing the concept of spatial regularity with graphs. We propose an efficient algorithm that tackles a broad family of spatial problems and provides novel convergence guarantees and significant speed-ups compared to generic approaches. The second chapter introduces a deep learning method that extends the idea of exploiting graph regularity to the case of massive 3D point clouds. We simplify the task of large-scale semantic segmentation by formulating it as as a small graph labelling problem. Our compact models reach high precision at a fraction of the computational cost of other approaches. In the third chapter, we present a collection of methods designed to take advantage of the data structure inherited from 3D sensors. By considering the sensors’ structure, we develop powerful networks with state-of-the-art accuracy, latency, and robustness for various applications and data types. The last chapter dives into the real-life challenge of automated satellite time series analysis for crop mapping. Recognizing the difference between such data and standard formats used in computer vision, we propose novel and streamlined architectures that achieve unprecedented precision while remaining efficient and economical in memory and preprocessing. We also introduce the task of panoptic segmentation for satellite time series and an efficient architecture to solve this problem at scale. In summary, this manuscript argues that geospatial problems represent a challenging and impactful venue for evaluating the newest machine learning and vision methods and a fertile source of inspiration for designing novel approaches. Note de contenu : 1- Introduction
2- Exploiting graph regularity
3- Exploiting the spatial regularity of 3D data
4- Exploiting the structure of 3D sensors
5- Exploiting the structure of satellite time series
6- Perspectives
7- Curriculum vitaeNuméro de notice : 24107 Affiliation des auteurs : UGE-LASTIG (2020- ) Thématique : IMAGERIE Nature : HDR Note de thèse : HDR: Sciences et Technologies de l’Information Geographique : UGE : 2023 Organisme de stage : LASTIG (IGN) DOI : sans En ligne : https://hal.science/tel-04095452v1 Format de la ressource électronique : URL Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=103248 A survey and benchmark of automatic surface reconstruction from point clouds / Raphaël Sulzer (2023)
![]()
Titre : A survey and benchmark of automatic surface reconstruction from point clouds Type de document : Article/Communication Auteurs : Raphaël Sulzer , Auteur ; Loïc Landrieu
, Auteur ; Renaud Marlet, Auteur ; Bruno Vallet
, Auteur
Editeur : Ithaca [New York - Etats-Unis] : ArXiv - Université Cornell Année de publication : 2023 Projets : BIOM / Vallet, Bruno Importance : 24 p. Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image
[Termes IGN] benchmark spatial
[Termes IGN] données d'entrainement (apprentissage automatique)
[Termes IGN] données lidar
[Termes IGN] données localisées 3D
[Termes IGN] reconstruction d'objet
[Termes IGN] semis de pointsRésumé : (auteur) We survey and benchmark traditional and novel learning-based algorithms that address the problem of surface reconstruction from point clouds. Surface reconstruction from point clouds is particularly challenging when applied to real-world acquisitions, due to noise, outliers, non-uniform sampling and missing data. Traditionally, different handcrafted priors of the input points or the output surface have been proposed to make the problem more tractable. However, hyperparameter tuning for adjusting priors to different acquisition defects can be a tedious task. To this end, the deep learning community has recently addressed the surface reconstruction problem. In contrast to traditional approaches, deep surface reconstruction methods can learn priors directly from a training set of point clouds and corresponding true surfaces. In our survey, we detail how different handcrafted and learned priors affect the robustness of methods to defect-laden input and their capability to generate geometric and topologically accurate reconstructions. In our benchmark, we evaluate the reconstructions of several traditional and learning-based methods on the same grounds. We show that learning-based methods can generalize to unseen shape categories, but their training and test sets must share the same point cloud characteristics. We also provide the code and data to compete in our benchmark and to further stimulate the development of learning-based surface reconstruction: https://github.com/raphaelsulzer/dsr-benchmark. Numéro de notice : P2023-004 Affiliation des auteurs : UGE-LASTIG+Ext (2020- ) Autre URL associée : vers ArXiv Thématique : IMAGERIE/INFORMATIQUE Nature : Preprint nature-HAL : Préprint DOI : 10.48550/arXiv.2301.13656 Date de publication en ligne : 31/01/2023 En ligne : https://hal.science/hal-03968453 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=102847 Where am I now? modelling disorientation in pan-scalar maps / Guillaume Touya in ISPRS International journal of geo-information, vol 12 n° 2 (February 2023)
PermalinkPermalink3D geovisualization for visual analysis of urban climate / Sidonie Christophe in Cybergeo, European journal of geography, vol 2022 ([01/01/2022])
PermalinkPermalinkPermalinkPermalinkA l'aide ! Je me suis perdu en zoomant / Guillaume Touya in Cartes & Géomatique, n° 247-248 (mars-juin 2022)
PermalinkALEGORIA: Joint multimodal search and spatial navigation into the geographic iconographic heritage / Florent Geniet (2022)
PermalinkPermalinkAnalysis of massive imports of open data in Openstreetmap database: a study case for France / Arnaud Le Guilcher in ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, vol V-4-2022 (2022 edition)
Permalink