Centre de documentation
scientifique

Dynamic scene understanding using deep neural networks / Ye Lyu (2021)  

Public Titre : Dynamic scene understanding using deep neural networks Type de document : Thèse/HDR Auteurs : Ye Lyu, Auteur ; M. George Vosselman, Directeur de thèse ; Michael Ying Yang, Directeur de thèse Editeur : Enschede [Pays-Bas] : International Institute for Geo-Information Science and Earth Observation ITC Année de publication : 2021 Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image [Termes IGN] attention (apprentissage automatique) [Termes IGN] chaîne de traitement [Termes IGN] champ aléatoire conditionnel [Termes IGN] compréhension de l'image [Termes IGN] détection d'objet [Termes IGN] image captée par drone [Termes IGN] image vidéo [Termes IGN] poursuite de cible [Termes IGN] régression [Termes IGN] segmentation sémantique Résumé : (auteur) Scene understanding is an important and fundamental research field in computer vision, which is quite useful for many applications in photogrammetry and remote sensing. It focuses on locating and classifying objects in images, understanding the relationships between them. The higher goal is to interpret what event happens in the scene, when it happens and why it happens, and what should we do based on the information. Dynamic scene understanding is to use information from different time to interpret scenes and answer the above related questions. For modern scene understanding technology, deep learning has shown great potential for such task. "Deep" in deep learning refers to the use of multiple layers in the neural networks. Deep neural networks are powerful as they are highly non-linear function that possess the ability to map from one domain to another quite different domain after proper training. It is the best solution for many fundamental research tasks regarding scene understanding. This ph.D. research also takes advantage of deep learning for dynamic scene understanding. Temporal information plays an important role for dynamic scene understanding. Compared with static scene understanding from images, information distilled from the time dimension provides values in many different ways. Images across consecutive frames have very high correlation, i.e., objects observed in one frame have very high chance to be observed and identified in nearby frames as well. Such redundancy in observation could potentially reduce the uncertainty for object recognition with deep learning based methods, resulting in more consistent inference. High correlation across frames could also improve the chance for recognizing objects correctly. If the camera or the object moves, the object could be observed in multiple different views with different poses and appearance. The information captured for object recognition would be more diverse and complementary, which could be aggregated to jointly inference the categories and the properties of objects. This ph.D. research involves several tasks related to the dynamic scene understanding in computer vision, including semantic segmentation for aerial platform images (chapter 2, 3), video object segmentation and video object detection for common objects in natural scenes (chapter 4, 5), and multi-object tracking and segmentation for cars and pedestrians in driving scenes (chapter 6). Chapter2 investigates how to establish the semantic segmentation benchmark for the UAV images, which includes data collection, data labeling, dataset construction, and performance evaluation with baseline deep neural networks and the proposed multi-scale dilation net. Conditional random field with feature space optimization is used to achieve consistent semantic segmentation prediction in videos. Chapter3 investigates how to better extract the scene context information for etter object recognition performance by proposing the novel bidirectional multiscale attention networks. It achieves better performance by inferring features and attention weights for feature fusing from both higher level and lower level branches. Chapter4 investigates how to simultaneously segment multiple objects across multiple frames by combining memory modules with instance segmentation networks. Our method learns to propagate the target object labels without auxiliary data, such as optical flow, which simplifies the model. Chapter5 investigates how to improve the performance of well-trained object detectors with a light weighted and efficient plug&play tracker for object detection in video. This chapter also investigates how the proposed model performs when lacking video training data. Chapter6 investigates how to improve the performance of detection, segmentation, and tracking by jointly considering top-down and bottom-up inference. The whole pipeline follows the multi-task design, i.e., a single feature extraction backbone with multiple heads for different sub-tasks. Overall, this manuscript has delved into several different computer vision tasks, which share fundamental research problems, including detection, segmentation, and tracking. Based on the research experiments and knowledge from literature review, several reflections regarding dynamic scene understanding have been discussed: The range of object context influence the quality for object recognition; The quality of video data affect the method choice for specific computer vision task; Detection and tracking are complementary for each other. For future work, unified dynamic scene understanding task could be a trend, and transformer plus self-supervised learning is one promising research direction. Real-time processing for dynamic scene understanding requires further researches in order to put the methods into usage for real-world applications. Numéro de notice : 12984 Affiliation des auteurs : non IGN Thématique : IMAGERIE/INFORMATIQUE Nature : Thèse étrangère Note de thèse : PhD thesis : Geo-Information Science and Earth Observation : Enschede, university of Twente : 2021 DOI : 10.3990/1.9789036552233 Date de publication en ligne : 08/09/2021 En ligne : https://library.itc.utwente.nl/papers_2021/phd/lyu.pdf Format de la ressource électronique : URL Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=100962

Geospatial analysis of the spreading of COVID-19 In the United States / Otto Heimonen (2021)  

Public Titre : Geospatial analysis of the spreading of COVID-19 In the United States Type de document : Mémoire Auteurs : Otto Heimonen, Auteur Editeur : Tampere [Finlande] : Tampere University Année de publication : 2021 Importance : 67 p. Format : 21 x 30 cm Note générale : bibliographie Master’s Degree Programme in Computational Big Data Analytics Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Analyse spatiale [Termes IGN] autocorrélation spatiale [Termes IGN] champ aléatoire conditionnel [Termes IGN] épidémie [Termes IGN] estimation bayesienne [Termes IGN] Etats-Unis [Termes IGN] maladie infectieuse [Termes IGN] méthode de Monte-Carlo par chaînes de Markov [Termes IGN] méthode du maximum de vraisemblance (estimation) [Termes IGN] modèle de simulation Résumé : (auteur) The COVID-19 pandemic has been a big threat to public health and there is an increasing need for efficient modelling of pathogens, predicting the daily infection rates to reduce the spread of COVID-19. The Moran’s and Geary’s statistics showed significant spatial autocorrelation in the infection counts for the US COVID-19 data. Spatial regression using the simultaneous autoregression (SAR) and conditional autoregression (CAR) models indicate clear association between the confirmed cases and the number of population and the population density in both national county and state specific analyses. The SAR model provided a better model fit with the low AIC value, leaving no significant autocorrelation for the residuals. The approximate Bayesian computation (ABC) methods were used to provide a flexible posterior distribution of the infection rate for COVID-19 based on the first 100 days of the pandemic. Three different simulation methods such as ABC-Rejection, ABC-Markov Chain Monte Carlo (MCMC) and ABC-Sequential Monte Carlo (SMC) were employed and compared. These algorithms seem to give reasonable posterior estimates for the average daily infections when the likelihood calculations for the spread of a harmful pathogen become complex, or intractable entirely. The posterior distributions of ABC-MCMC and ABC-SMC provided plausible estimations covering all of the observed infection rates at different time points. Note de contenu : 1- Introduction 2- Methods 3- Empirical data analysis 4- Discussion Numéro de notice : 28455 Affiliation des auteurs : non IGN Thématique : GEOMATIQUE/MATHEMATIQUE Nature : Mémoire masters divers DOI : sans En ligne : https://trepo.tuni.fi/handle/10024/134567 Format de la ressource électronique : URL Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=99025

Addressing overfitting on point cloud classification using Atrous XCRF / Hasan Asy’ari Arief in ISPRS Journal of photogrammetry and remote sensing, vol 155 (September 2019)  

Public [article]  inISPRS Journal of photogrammetry and remote sensing > vol 155 (September 2019) . - pp 90 - 101 Titre : Addressing overfitting on point cloud classification using Atrous XCRF Type de document : Article/Communication Auteurs : Hasan Asy’ari Arief, Auteur ; Ulf Geir Indahl, Auteur ; Geir-Harald Strand, Auteur ; Håvard Tveite, Auteur Année de publication : 2019 Article en page(s) : pp 90 - 101 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Lasergrammétrie [Termes IGN] champ aléatoire conditionnel [Termes IGN] classification automatique [Termes IGN] réseau neuronal convolutif [Termes IGN] réseau neuronal profond [Termes IGN] semis de points Résumé : (Auteur) Advances in techniques for automated classification of point cloud data introduce great opportunities for many new and existing applications. However, with a limited number of labelled points, automated classification by a machine learning model is prone to overfitting and poor generalization. The present paper addresses this problem by inducing controlled noise (on a trained model) generated by invoking conditional random field similarity penalties using nearby features. The method is called Atrous XCRF and works by forcing a trained model to respect the similarity penalties provided by unlabeled data. In a benchmark study carried out using the ISPRS 3D labeling dataset, our technique achieves 85.0% in term of overall accuracy, and 71.1% in term of F1 score. The result is on par with the current best model for the benchmark dataset and has the highest value in term of F1 score. Additionally, transfer learning using the Bergen 2018 dataset, without model retraining, was also performed. Even though our proposal provides a consistent 3% improvement in term of accuracy, more work still needs to be done to alleviate the generalization problem on the domain adaptation and the transfer learning field. Numéro de notice : A2019-312 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article DOI : 10.1016/j.isprsjprs.2019.07.002 Date de publication en ligne : 11/07/2019 En ligne : https://doi.org/10.1016/j.isprsjprs.2019.07.002 Format de la ressource électronique : URL Article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=93337 [article]

Exemplaires(3)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
081-2019091	RAB	Revue	Centre de documentation	En réserve L003	Disponible
081-2019093	DEP-RECP	Revue	LASTIG	Dépôt en unité	Exclu du prêt
081-2019092	DEP-RECF	Revue	Nancy	Dépôt en unité	Exclu du prêt

Structural segmentation and classification of mobile laser scanning point clouds with large variations in point density / Yuan Li in ISPRS Journal of photogrammetry and remote sensing, vol 153 (July 2019)  

Public [article]  inISPRS Journal of photogrammetry and remote sensing > vol 153 (July 2019) . - pp 151 - 165 Titre : Structural segmentation and classification of mobile laser scanning point clouds with large variations in point density Type de document : Article/Communication Auteurs : Yuan Li, Auteur ; Bo Wu, Auteur ; Xuming Ge, Auteur Année de publication : 2019 Article en page(s) : pp 151 - 165 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Lasergrammétrie [Termes IGN] champ aléatoire conditionnel [Termes IGN] classification [Termes IGN] classification basée sur les régions [Termes IGN] densité des points [Termes IGN] données lidar [Termes IGN] données localisées 3D [Termes IGN] Hong-Kong [Termes IGN] modèle 3D de l'espace urbain [Termes IGN] Paris (75) [Termes IGN] scène urbaine [Termes IGN] segmentation en régions [Termes IGN] segmentation hiérarchique [Termes IGN] segmentation sémantique [Termes IGN] semis de points Résumé : (Auteur) Objects are formed by various structures and such structural information is essential for the identification of objects, especially for street facilities presented by mobile laser scanning (MLS) data with abundant details. However, due to the large volume of data, large variations in point density, noise and complexity of scanned scenes, the achievement of effective decomposition of objects into physical meaningful structures remains a challenge issue. And structural information has been rarely considered to improve the accuracy of distinguishing between objects with global or local similarity, such as traffic signs and traffic lights. Therefore, we propose a structural segmentation and classification method for MLS point clouds that is efficient and robust to variations in point density and complex urban scenes. During the segmentation stage, a novel region growing approach and a multi-size supervoxel segmentation algorithm robust to noise and varying density are combined to extract effective local shape descriptors. Structural components with physically meaningful labels are generated via structural labelling and clustering. During the classification stage, we consider the structural information at various scales and locations and encode it into a conditional random-field model for unary and pairwise inferences. High-order potentials are also introduced into the conditional random field to eliminate regional label noise. These high-order potentials are defined upon regions independent of connection relationships and can therefore take effect on isolated nodes. Experiments with two MLS datasets of typical urban scenes in Paris and Hong Kong were used to evaluate the performance of the proposed method. Nine and eleven different object classes were recognized from these two datasets with overall accuracies of 97.13% and 95.79%, respectively, indicating the effectiveness of the proposed method of interpreting complex urban scenes from point clouds with large variations in point density. Compared with previous studies on the Paris dataset, our method was able to recognize more classes and obtained a mean F1-score of 72.70% of seven common classes, being higher than the best of previous results. Numéro de notice : A2019-262 Affiliation des auteurs : non IGN Thématique : IMAGERIE/URBANISME Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1016/j.isprsjprs.2019.05.007 Date de publication en ligne : 28/05/2019 En ligne : https://doi.org/10.1016/j.isprsjprs.2019.05.007 Format de la ressource électronique : URL Article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=93075 [article]

Exemplaires(3)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
081-2019071	RAB	Revue	Centre de documentation	En réserve L003	Disponible
081-2019073	DEP-RECP	Revue	LASTIG	Dépôt en unité	Exclu du prêt
081-2019072	DEP-RECF	Revue	Nancy	Dépôt en unité	Exclu du prêt

Semantic façade segmentation from airborne oblique images / Yaping Lin in Photogrammetric Engineering & Remote Sensing, PERS, vol 85 n° 6 (June 2019)  

Public [article]  inPhotogrammetric Engineering & Remote Sensing, PERS > vol 85 n° 6 (June 2019) . - pp 425 - 433 Titre : Semantic façade segmentation from airborne oblique images Type de document : Article/Communication Auteurs : Yaping Lin, Auteur ; Francesco Nex, Auteur ; Michael Ying Yang, Auteur Année de publication : 2019 Article en page(s) : pp 425 - 433 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] analyse comparative [Termes IGN] champ aléatoire conditionnel [Termes IGN] classification par forêts d'arbres décisionnels [Termes IGN] classification par réseau neuronal convolutif [Termes IGN] façade [Termes IGN] image aérienne oblique [Termes IGN] image RVB [Termes IGN] segmentation d'image [Termes IGN] segmentation sémantique [Termes IGN] semis de points Résumé : (Auteur) In this paper, oblique airborne images with very high resolution are used to address the problem from aerial views in urban areas. Traditional classification method (i.e., random forests) is compared with state-of-the-art fully convolutional networks (FCNs). Random forests use hand-craft image features including red, green, blue (RGB), scale-invariant feature transform (SIFT), and Texton, and point cloud features consisting of normal vector and planarity extracted from different scales. In contrast, the inputs of FCNs are the RGB bands and the third components of normal vectors. In both cases, three-dimensional (3D) features are projected back into the image space to support the facade interpretation. Fully connected conditional random field (CRF) is finally taken as a post-processing of the FCN to refine the segmentation results. Several tests have been performed and the achieved results show that the models embedding the 3D component outperform the solution using only images. FCNs significantly outperformed random forests, especially for the balcony delineation. Numéro de notice : A2019-247 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.14358/PERS.85.6.425 Date de publication en ligne : 01/06/2019 En ligne : https://doi.org/10.14358/PERS.85.6.425 Format de la ressource électronique : URL Article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=93003 [article]

Exemplaires(1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
105-2019061	SL	Revue	Centre de documentation	Revues en salle	Disponible

Conditional random field and deep feature learning for hyperspectral image classification / Fahim Irfan Alam in IEEE Transactions on geoscience and remote sensing, vol 57 n° 3 (March 2019)  

Permalink

Automatic building rooftop extraction from aerial images via hierarchical RGB-D priors / Shibiao Xu in IEEE Transactions on geoscience and remote sensing, vol 56 n° 12 (December 2018)  

Permalink

Deep multi-task learning for a geographically-regularized semantic segmentation of aerial images / Michele Volpi in ISPRS Journal of photogrammetry and remote sensing, vol 144 (October 2018)  

Permalink

Contextual classification using photometry and elevation data for damage detection after an earthquake event / Ewelina Rupnik in European journal of remote sensing, vol 51 n° 1 (2018)  

Permalink

Crop-rotation structured classification using multi-source sentinel images and LPIS for crop type mapping / Simon Bailly (2018)  

Permalink

Learning aggregated features and optimizing model for semantic labeling / Jianhua Wang in The Visual Computer, vol 33 n° 12 (December 2017)  

Permalink

A higher order conditional random field model for simultaneous classification of land cover and land use / Lena Albert in ISPRS Journal of photogrammetry and remote sensing, vol 130 (August 2017)  

Permalink

Analyse de séries temporelles d’images Sentinel et intégration de connaissances pour la classification en milieu agricole / Simon Bailly (2017) 

Permalink

Comparison of belief propagation and graph-cut approaches for contextual classification of 3D LIDAR point cloud data / Loïc Landrieu (2017)    

Permalink

Pré-segmentation pour la classification faiblement supervisée de scènes urbaines à partir de nuages de points 3D LIDAR / Stéphane Guinard (2017)  

Permalink