Descripteur
|
Termes descripteurs IGN > informatique > intelligence artificielle > vision par ordinateur
vision par ordinateur |
Ajouter le résultat dans votre panier
Visionner les documents numériques
Affiner la recherche Interroger des sources externesEtendre la recherche sur niveau(x) vers le bas
Context pyramidal network for stereo matching regularized by disparity gradients / Junhua Kang in ISPRS Journal of photogrammetry and remote sensing, vol 157 (November 2019)
Titre : Context pyramidal network for stereo matching regularized by disparity gradients Type de document : Article/Communication Auteurs : Junhua Kang, Auteur ; Lin Chen, Auteur ; Fei Deng, Auteur ; Christian Heipke, Auteur Année de publication : 2019 Article en page(s) : pp 201 - 215 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Termes descripteurs IGN] appariement d'images
[Termes descripteurs IGN] appariement de formes
[Termes descripteurs IGN] apprentissage profond
[Termes descripteurs IGN] chaîne de traitement
[Termes descripteurs IGN] classification par réseau neuronal convolutif
[Termes descripteurs IGN] extraction de traits caractéristiques
[Termes descripteurs IGN] gradient
[Termes descripteurs IGN] vision par ordinateur
[Termes descripteurs IGN] vision stéréoscopiqueRésumé : (Auteur) Also after many years of research, stereo matching remains to be a challenging task in photogrammetry and computer vision. Recent work has achieved great progress by formulating dense stereo matching as a pixel-wise learning task to be resolved with a deep convolutional neural network (CNN). However, most estimation methods, including traditional and deep learning approaches, still have difficulty to handle real-world challenging scenarios, especially those including large depth discontinuity and low texture areas.
To tackle these problems, we investigate a recently proposed end-to-end disparity learning network, DispNet (Mayer et al., 2015), and improve it to yield better results in these problematic areas. The improvements consist of three major contributions. First, we use dilated convolutions to develop a context pyramidal feature extraction module. A dilated convolution expands the receptive field of view when extracting features, and aggregates more contextual information, which allows our network to be more robust in weakly textured areas. Second, we construct the matching cost volume with patch-based correlation to handle larger disparities. We also modify the basic encoder-decoder module to regress detailed disparity images with full resolution. Third, instead of using post-processing steps to impose smoothness in the presence of depth discontinuities, we incorporate disparity gradient information as a gradient regularizer into the loss function to preserve local structure details in large depth discontinuity areas.
We evaluate our model in terms of end-point-error on several challenging stereo datasets including Scene Flow, Sintel and KITTI. Experimental results demonstrate that our model decreases the estimation error compared with DispNet on most datasets (e.g. we obtain an improvement of 46% on Sintel) and estimates better structure-preserving disparity maps. Moreover, our proposal also achieves competitive performance compared to other methods.Numéro de notice : A2019-496 Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1016/j.isprsjprs.2019.09.012 date de publication en ligne : 27/09/2019 En ligne : https://doi.org/10.1016/j.isprsjprs.2019.09.012 Format de la ressource électronique : URL Article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=93729
in ISPRS Journal of photogrammetry and remote sensing > vol 157 (November 2019) . - pp 201 - 215[article]Réservation
Réserver ce documentExemplaires (3)
Code-barres Cote Support Localisation Section Disponibilité 081-2019111 SL Revue Centre de documentation Revues en salle Disponible 081-2019113 DEP-RECP Revue MATIS Dépôt en unité Exclu du prêt 081-2019112 DEP-RECF Revue Nancy Dépôt en unité Exclu du prêt
Titre : Detecting and mapping traffic signs from Google Street View images using deep learning and GIS Type de document : Article/Communication Auteurs : Andrew Campbell, Auteur ; Alan Both, Auteur ; Qian (Chayn) Sun, Auteur ; Qian (Chayn) Sun, Auteur Année de publication : 2019 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Applications SIG
[Termes descripteurs IGN] apprentissage profond
[Termes descripteurs IGN] base de données routières
[Termes descripteurs IGN] détection d'objet
[Termes descripteurs IGN] feu de circulation
[Termes descripteurs IGN] gestion de trafic
[Termes descripteurs IGN] image Streetview
[Termes descripteurs IGN] réseau routier
[Termes descripteurs IGN] signalisation routière
[Termes descripteurs IGN] système d'information géographique
[Termes descripteurs IGN] trafic routier
[Termes descripteurs IGN] vision par ordinateurRésumé : (auteur) Street traffic sign infrastructure remains an extremely difficult asset for local government to manage due to its diverse physical structure and geographical distribution. A spatial registrar of traffic infrastructure is currently a required component of local government councils' mandatory road management plans. Recent advancements of object detection technology in machine learning have presented an automated approach for the detection and classification of street signage captured by Google's Street View (GSV) imagery. This paper explores the possibility of using deep learning to produce an autonomous system for detecting traffic signs on GSV images to assist in traffic assets monitoring and maintenance. By leveraging Google's Street View API, this research offers an economic approach of building purposeful street sign computer vision datasets. A custom object detection model was trained to detect and classify Stop and Give Way signs from images captured at intersection approaches. Considering the output detected bounding box coordinates, photogrammetry approach was applied to calculate the approximate location of each detected sign in two-dimensional geographical space. The newly located and classified street signs can be combined with relevant spatial data for implementation into an asset management system. By combining GIS and the GSV API, the process is completely scalable to any level of street sign classification scope. The experiments conducted on the road network of study area recorded a detection accuracy of 95.63% and classification accuracy of 97.82%. Our proposed automated approach to the detection and localisation of street sign infrastructure has displayed a promising potential for its use by local government authorities. Our workflow can be used to detect other traffic signs and applied to other road sections and other cities. Of primary importance, this approach takes an entirely free and open-source approach throughout. The continuation of Google's Street View program will account for the spatiotemporal representation of street sign infrastructure for the ongoing maintenance and renewal programs of this valuable asset. Numéro de notice : A2019-412 Thématique : GEOMATIQUE/IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1016/j.compenvurbsys.2019.101350 date de publication en ligne : 07/06/2019 En ligne : https://doi.org/10.1016/j.compenvurbsys.2019.101350 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=93536
in Computers, Environment and Urban Systems > vol 77 (september 2019)[article]
Titre : Learning scene geometry for visual localization in challenging conditions Type de document : Article/Communication Auteurs : Nathan Piasco Congrès : ICRA 2019, International Conference on Robotics and Automation (20 - 24 mai 2019; Montréal, Québec - Canada) Editeur : New York : Institute of Electrical and Electronics Engineers IEEE Année de publication : 2019 Projets : PLaTINUM / Gouet-Brunet, Valérie Importance : pp 9094 - 9100 Format : 21 x 30 cm Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique
[Termes descripteurs IGN] analyse d'image orientée objet
[Termes descripteurs IGN] analyse visuelle
[Termes descripteurs IGN] appariement d'images
[Termes descripteurs IGN] carte de profondeur
[Termes descripteurs IGN] descripteur
[Termes descripteurs IGN] géométrie de l'image
[Termes descripteurs IGN] image RVB
[Termes descripteurs IGN] localisation basée vision
[Termes descripteurs IGN] précision de localisation
[Termes descripteurs IGN] prise de vue nocturne
[Termes descripteurs IGN] robotique
[Termes descripteurs IGN] scène urbaine
[Termes descripteurs IGN] variation diurne
[Termes descripteurs IGN] variation saisonnière
[Termes descripteurs IGN] vision par ordinateurRésumé : (auteur) We propose a new approach for outdoor large scale image based localization that can deal with challenging scenarios like cross-season, cross-weather, day/night and longterm localization. The key component of our method is a new learned global image descriptor, that can effectively benefit from scene geometry information during training. At test time, our system is capable of inferring the depth map related to the query image and use it to increase localization accuracy. We are able to increase recall@1 performances by 2.15% on cross-weather and long-term localization scenario and by 4.24% points on a challenging winter/summer localization sequence versus state-of-the-art methods. Our method can also use weakly annotated data to localize night images across a reference dataset of daytime images. Numéro de notice : C2019-002 Thématique : IMAGERIE Nature : Communication nature-HAL : ComAvecCL&ActesPubliésIntl DOI : 10.1109/ICRA.2019.8794221 date de publication en ligne : 12/08/2019 En ligne : http://doi.org/10.1109/ICRA.2019.8794221 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=93774 Documents numériques
en open access
Learning scene geometry... - pdf auteurAdobe Acrobat PDF
Titre : Structure from motion for ordered and unordered image sets based on random k-d forests and global pose estimation Type de document : Article/Communication Auteurs : Xin Wang, Auteur ; Franz Rottensteiner, Auteur ; Christian Heipke, Auteur Année de publication : 2019 Article en page(s) : pp 19 - 41 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image
[Termes descripteurs IGN] appariement d'images
[Termes descripteurs IGN] chaîne de traitement
[Termes descripteurs IGN] classification barycentrique
[Termes descripteurs IGN] classification par forêts aléatoires
[Termes descripteurs IGN] compensation par faisceaux
[Termes descripteurs IGN] estimation de pose
[Termes descripteurs IGN] image de drone
[Termes descripteurs IGN] matrice de rotation
[Termes descripteurs IGN] orientation relative
[Termes descripteurs IGN] Ransac (algorithme)
[Termes descripteurs IGN] recouvrement d'images
[Termes descripteurs IGN] SIFT (algorithme)
[Termes descripteurs IGN] structure-from-motion
[Termes descripteurs IGN] vision par ordinateurRésumé : (auteur) In this paper, we present a new fast and robust method for structure from motion (SfM) for data sets potentially comprising thousands of ordered or unordered images. Our work focuses on the two most time-consuming procedures: (a) image matching and (b) pose estimation. For image matching, a new method employing a random k-d forest is proposed to quickly obtain pairs of overlapping images from an unordered set. After that, image matching and the estimation of relative orientation parameters are performed only for pairs found to be very likely to overlap. For pose estimation, we use a two-stage global approach, separating the determination of rotation matrices and translation parameters; the latter are computed simultaneously using a new method. In order to cope with outliers in the relative orientations, which global approaches are particularly sensitive to, we present a new constraint based on triplet loop closure errors of rotation and translation. Finally, a robust bundle adjustment is carried out to refine the image orientation parameters. We demonstrate the potential and limitations of our pipeline using various real-world datasets including ordered image data acquired from UAV (unmanned aerial vehicle) and other platforms as well as unordered data from the internet. The experiments show that our work performs better than comparable state-of-the-art SfM systems in terms of run time, while we achieve a similar accuracy and robustness. Numéro de notice : A2019-033 Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1016/j.isprsjprs.2018.11.009 date de publication en ligne : 15/11/2018 En ligne : https://doi.org/10.1016/j.isprsjprs.2018.11.009 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=91970
in ISPRS Journal of photogrammetry and remote sensing > vol 147 (January 2019) . - pp 19 - 41[article]Réservation
Réserver ce documentExemplaires (3)
Code-barres Cote Support Localisation Section Disponibilité 081-2019011 SL Revue Centre de documentation Revues en salle Disponible 081-2019013 DEP-EXM Revue MATIS Dépôt en unité Exclu du prêt 081-2019012 DEP-EAF Revue Nancy Dépôt en unité Exclu du prêt
Titre : Augmented reality meets computer vision : efficient data generation for urban driving scenes Type de document : Article/Communication Auteurs : Hassan Abu Alhaija, Auteur ; Siva Karthik Mustikovela, Auteur ; Lars Mescheder, Auteur ; Andreas Geiger, Auteur ; Carsten Rother, Auteur Année de publication : 2018 Article en page(s) : pp 961 - 972 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image
[Termes descripteurs IGN] apprentissage automatique
[Termes descripteurs IGN] détection d'objet
[Termes descripteurs IGN] réalité augmentée
[Termes descripteurs IGN] scène urbaine
[Termes descripteurs IGN] vision par ordinateurRésumé : (Auteur) The success of deep learning in computer vision is based on the availability of large annotated datasets. To lower the need for hand labeled images, virtually rendered 3D worlds have recently gained popularity. Unfortunately, creating realistic 3D content is challenging on its own and requires significant human effort. In this work, we propose an alternative paradigm which combines real and synthetic data for learning semantic instance segmentation and object detection models. Exploiting the fact that not all aspects of the scene are equally important for this task, we propose to augment real-world imagery with virtual objects of the target category. Capturing real-world images at large scale is easy and cheap, and directly provides real background appearances without the need for creating complex 3D models of the environment. We present an efficient procedure to augment these images with virtual objects. In contrast to modeling complete 3D environments, our data augmentation approach requires only a few user interactions in combination with 3D models of the target object category. Leveraging our approach, we introduce a novel dataset of augmented urban driving scenes with 360 degree images that are used as environment maps to create realistic lighting and reflections on rendered objects. We analyze the significance of realistic object placement by comparing manual placement by humans to automatic methods based on semantic scene analysis. This allows us to create composite images which exhibit both realistic background appearance as well as a large number of complex object arrangements. Through an extensive set of experiments, we conclude the right set of parameters to produce augmented data which can maximally enhance the performance of instance segmentation models. Further, we demonstrate the utility of the proposed approach on training standard deep models for semantic instance segmentation and object detection of cars in outdoor driving scenarios. We test the models trained on our augmented data on the KITTI 2015 dataset, which we have annotated with pixel-accurate ground truth, and on the Cityscapes dataset. Our experiments demonstrate that the models trained on augmented imagery generalize better than those trained on fully synthetic data or models trained on limited amounts of annotated real data. Numéro de notice : A2018-417 Thématique : IMAGERIE/INFORMATIQUE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1007/s11263-018-1070-x date de publication en ligne : 07/03/2018 En ligne : https://doi.org/10.1007/s11263-018-1070-x Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=90900
in International journal of computer vision > vol 126 n° 9 (September 2018) . - pp 961 - 972[article]Permalink
PermalinkPermalinkLocalisation par l'image en milieu urbain : application à la réalité augmentée / Antoine Fond (2018)
PermalinkMachine learning and pose estimation for autonomous robot grasping with collaborative robots / Victor Talbot (2018)
PermalinkPermalinkPermalinkPermalinkORASIS 2017, 16e journées francophones des jeunes chercheurs en vision par ordinateur, 12 - 16 juin 2017 au Club Belambra “Omaha Beach” Le Cavey à Colleville-sur-Mer (Normandie) / Laboratoire Groupe de Recherche en Informatique, Image, Automatique et Instrumentation de Caen (Campus 2 - Pôle Sciences et technologies de l’Université de Caen, Caen, France) (2017)
PermalinkPermalink
