Centre de documentation
scientifique

GeoMultiTaskNet: remote sensing unsupervised domain adaptation using geographical coordinates / Valerio Marsocci (2023)  

Public Titre : GeoMultiTaskNet: remote sensing unsupervised domain adaptation using geographical coordinates Type de document : Article/Communication Auteurs : Valerio Marsocci, Auteur ; Nicolas Gonthier, Auteur ; Anatol Garioud , Auteur ; Simone Scardapane, Auteur ; Clément Mallet , Auteur Editeur : Ithaca [New York - Etats-Unis] : ArXiv - Université Cornell 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 : 11 p. Format : 21 x 30 cm Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] base de données d'occupation du sol [Termes IGN] image à très haute résolution [Termes IGN] jeu de données localisées [Termes IGN] métadonnées géographiques [Termes IGN] segmentation sémantique Résumé : (auteur) Land cover maps are a pivotal element in a wide range of Earth Observation (EO) applications. However, annotating large datasets to develop supervised systems for remote sensing (RS) semantic segmentation is costly and time-consuming. Unsupervised Domain Adaption (UDA) could tackle these issues by adapting a model trained on a source domain, where labels are available, to a target domain, without annotations. UDA, while gaining importance in computer vision, is still under-investigated in RS. Thus, we propose a new lightweight model, GeoMultiTaskNet, based on two contributions: a GeoMultiTask module (GeoMT), which utilizes geographical coordinates to align the source and target domains, and a Dynamic Class Sampling (DCS) strategy, to adapt the semantic segmentation loss to the frequency of classes. This approach is the first to use geographical metadata for UDA in semantic segmentation. It reaches state-of-the-art performances (47,22% mIoU), reducing at the same time the number of parameters (33M), on a subset of the FLAIR dataset, a recently proposed dataset properly shaped for RS UDA, used for the first time ever for research scopes here. Numéro de notice : C2023-004 Affiliation des auteurs : UGE-LASTIG+Ext (2020- ) Autre URL associée : vers HAL Thématique : IMAGERIE Nature : Communication DOI : 10.48550/arXiv.2304.07750 En ligne : https://doi.org/10.48550/arXiv.2304.07750 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=103083

PSMNet-FusionX3 : LiDAR-guided deep learning stereo dense matching on aerial images / Teng Wu (2023)  

Public 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.pdf Langues : 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 Network Résumé : (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