Centre de documentation
scientifique

Pose estimation and 3D reconstruction of vehicles from stereo-images using a subcategory-aware shape prior / Maximilian Alexander Coenen in ISPRS Journal of photogrammetry and remote sensing, Vol 181 (November 2021)  

Public [article]  inISPRS Journal of photogrammetry and remote sensing > Vol 181 (November 2021) . - pp 27 - 47 Titre : Pose estimation and 3D reconstruction of vehicles from stereo-images using a subcategory-aware shape prior Type de document : Article/Communication Auteurs : Maximilian Alexander Coenen, Auteur ; Franz Rottensteiner, Auteur Année de publication : 2021 Article en page(s) : pp 27 - 47 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] classification par réseau neuronal convolutif [Termes IGN] détection d'objet [Termes IGN] estimation de pose [Termes IGN] modèle stochastique [Termes IGN] problème inverse [Termes IGN] reconstruction 3D [Termes IGN] reconstruction d'objet [Termes IGN] robotique [Termes IGN] véhicule automobile [Termes IGN] vision par ordinateur Résumé : (auteur) The 3D reconstruction of objects is a prerequisite for many highly relevant applications of computer vision such as mobile robotics or autonomous driving. To deal with the inverse problem of reconstructing 3D objects from their 2D projections, a common strategy is to incorporate prior object knowledge into the reconstruction approach by establishing a 3D model and aligning it to the 2D image plane. However, current approaches are limited due to inadequate shape priors and the insufficiency of the derived image observations for a reliable alignment with the 3D model. The goal of this paper is to show how 3D object reconstruction can profit from a more sophisticated shape prior and from a combined incorporation of different observation types inferred from the images. We introduce a subcategory-aware deformable vehicle model that makes use of a prediction of the vehicle type for a more appropriate regularisation of the vehicle shape. A multi-branch CNN is presented to derive predictions of the vehicle type and orientation. This information is also introduced as prior information for model fitting. Furthermore, the CNN extracts vehicle keypoints and wireframes, which are well-suited for model-to-image association and model fitting. The task of pose estimation and reconstruction is addressed by a versatile probabilistic model. Extensive experiments are conducted using two challenging real-world data sets on both of which the benefit of the developed shape prior can be shown. A comparison to state-of-the-art methods for vehicle pose estimation shows that the proposed approach performs on par or better, confirming the suitability of the developed shape prior and probabilistic model for vehicle reconstruction. Numéro de notice : A2021-772 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1016/j.isprsjprs.2021.07.006 Date de publication en ligne : 14/09/2021 En ligne : https://doi.org/10.1016/j.isprsjprs.2021.07.006 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=98829 [article]

Probabilistic pose estimation and 3D reconstruction of vehicles from stereo images / Maximilian Alexander Coenen (2020)  

Public Titre : Probabilistic pose estimation and 3D reconstruction of vehicles from stereo images Type de document : Thèse/HDR Auteurs : Maximilian Alexander Coenen, Auteur Editeur : Munich : Bayerische Akademie der Wissenschaften Année de publication : 2020 Collection : DGK - C, ISSN 0065-5325 num. 857 Importance : 160 p. ISBN/ISSN/EAN : 978-3-7696-5269-7 Note générale : bibliographie Diese Arbeit ist gleichzeitig veröffentlicht in: Wissenschaftliche Arbeiten der Fachrichtung Geodäsie und Geoinformatik der Universität Hannover Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Applications photogrammétriques [Termes IGN] classification par réseau neuronal convolutif [Termes IGN] estimation de pose [Termes IGN] méthode de Monte-Carlo [Termes IGN] modèle stochastique [Termes IGN] reconstruction 3D [Termes IGN] véhicule automobile Résumé : (auteur) The pose estimation and reconstruction of 3D objects from images is one of the major problems that are addressed in computer vision and photogrammetry. The understanding of a 3D scene and the 3D reconstruction of specific objects are prerequisites for many highly relevant applications of computer vision such as mobile robotics and autonomous driving. To deal with the inverse problem of reconstructing 3D objects from their 2D projections, a common strategy is to incorporate prior object knowledge into the reconstruction approach by establishing a 3D model and aligning it to the 2D image plane. However, current approaches are limited due to inadequate shape priors and the insufficiency of the derived image observations for a reliable association and alignment with the 3D model. The goal of this thesis is to infer valuable observations from the images and to show how 3D object reconstruction can profit from a more sophisticated shape prior and from a combined incorporation of the different observation types. To achieve this goal, this thesis presents three major contributions for the particular task of 3Dvehicle reconstruction from street-level stereo images. First, a subcategory-aware deformable vehicle model is introduced that makes use of a prediction of the vehicle type for a more appropriate regularisation of the vehicle shape. Second, a Convolutional Neural Network (CNN) is proposed which extracts observations from an image. In particular, the CNN is used to derive a prediction of the vehicle orientation and type, which are introduced as prior information for model fitting. Furthermore, the CNN extracts vehicle key points and wireframes, which are well-suited for model association and model fitting. Third, the task of pose estimation and reconstruction is addressed by a versatile probabilistic model. Suitable parametrisations and formulations of likelihood and prior terms are introduced for a joint consideration of the derived observations and prior information in the probabilistic objective function. As the objective function is non-convex and discontinuous, a proper customized strategy based on stochastic sampling is proposed for inference, yielding convincing results for the estimated poses and shapes of the vehicles. To evaluate the performance and to investigate the strengths and limitations of the proposed method, extensive experiments are conducted using two challenging real-world data sets: the publicly available KITTI benchmark and the ICSENS data set, which was created in the scope of this thesis. On both data sets, the benefit of the developed shape prior and of each of the individual components of the probabilistic model can be shown. The proposed method yields vehicle pose estimates with a median error of up to 27 cm for the position and up to 1.7◦for the orientation on the data sets. A comparison to state-of-the-art methods for vehicle pose estimation shows that the proposed approach performs on par or better, confirming the suitability of the developed model and inference procedure. Numéro de notice : 17685 Affiliation des auteurs : non IGN Autre URL associée : vers ResearchGate Thématique : IMAGERIE/INFORMATIQUE Nature : Thèse étrangère Note de thèse : PhD thesis : Geodäsie und Geoinformatik : Hanovre : 2020 DOI : 10.13140/RG.2.2.19618.86728 En ligne : https://dgk.badw.de/fileadmin/user_upload/Files/DGK/docs/c-857.pdf Format de la ressource électronique : URL Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=98165