pr\u00e9sentent<\/p>\n\n\n\n
Autoris\u00e9e \u00e0 pr\u00e9senter ses travaux en vue de l\u2019obtention du Doctorat de l’Institut Polytechnique de Paris, pr\u00e9par\u00e9 \u00e0 T\u00e9l\u00e9com SudParis en :<\/p>\n\n\n\n
le JEUDI 14 D\u00c9CEMBRE 2023 \u00e0 16h00<\/p>\n\n\n\n
\u00e0 T\u00e9l\u00e9com SudParis<\/p>\n\n\n\n
SALLE C06
9 Rue Charles Fourier, EVRY Courcouronnes FR, 91000<\/p>\n\n\n\n
Membres du jury :<\/strong><\/p>\n\n\n\n M. Titus ZAHARIA<\/strong>, Professor, T\u00e9l\u00e9com SudParis, FRANCE – Directeur de these R\u00e9sum\u00e9 :<\/strong><\/p>\n\n\n\n Dans l’Industrie 4.0, l’utilisation d’appareils de R\u00e9alit\u00e9 Augment\u00e9e (RA) tels que HoloLens a acquis une acceptation significative pour la formation des op\u00e9rateurs de ligne d’assemblage dans diverses industries. La simplification des mod\u00e8les CAO 3D pour la formation en RA est essentielle pour une meilleure performance des applications. Notre recherche se concentre sur le d\u00e9veloppement de m\u00e9thodes et de techniques visant \u00e0 rationaliser des mod\u00e8les CAO 3D complexes, les rendant adapt\u00e9s aux applications de RA. Notre recherche met en avant le r\u00f4le des mod\u00e8les 3D en RA, am\u00e9liorant l’exp\u00e9rience virtuelle en superposant des mod\u00e8les CAO sur le monde r\u00e9el. Nous d\u00e9taillons les applications de la RA dans la formation des op\u00e9rateurs et comment l’int\u00e9gration de mod\u00e8les CAO 3D am\u00e9liore la compr\u00e9hension des instructions et des proc\u00e9dures. Nous avons r\u00e9alis\u00e9 une revue approfondie de la litt\u00e9rature sur la simplification des mod\u00e8les CAO pour leur int\u00e9gration dans des sc\u00e9narios de r\u00e9alit\u00e9 augment\u00e9e (RA). Nos conclusions indiquent que les techniques de simplification bas\u00e9es sur le maillage excellent dans la pr\u00e9servation des \u00e9l\u00e9ments essentiels des mod\u00e8les CAO, offrant un contr\u00f4le pr\u00e9cis sur les niveaux de d\u00e9tail. De plus, nous avons effectu\u00e9 quatre types distincts d’\u00e9valuations dans notre recherche. Ces \u00e9valuations comprenaient des \u00e9valuations objectives utilisant des techniques bas\u00e9es sur le maillage issues de la litt\u00e9rature existante, des avis d’experts impliquant un examen approfondi de chaque mod\u00e8le simplifi\u00e9 pour d\u00e9terminer le niveau de simplification en fonction des plages de sommets, des tests en conditions r\u00e9elles assist\u00e9s par HoloLens2, qui ont r\u00e9v\u00e9l\u00e9 des am\u00e9liorations du taux de rafra\u00eechissement lors de l’utilisation de mod\u00e8les CAO au lieu de leurs versions originales. Pour conclure nos \u00e9valuations, nous avons \u00e9galement r\u00e9alis\u00e9 des \u00e9valuations par les utilisateurs, en donnant la priorit\u00e9 \u00e0 l’exp\u00e9rience utilisateur dans notre \u00e9tude. Ces \u00e9valuations ont confirm\u00e9 que les mod\u00e8les simplifi\u00e9s sont hautement capables de remplacer les versions originales. Cependant, il a \u00e9t\u00e9 observ\u00e9 qu’une simplification suppl\u00e9mentaire est n\u00e9cessaire, en particulier pour les mod\u00e8les CAO complexes. La m\u00e9thodologie principale propose une approche innovante ax\u00e9e sur la segmentation du maillage et la simplification adaptative gr\u00e2ce \u00e0 l’utilisation de m\u00e9thodes d’apprentissage profond. Pour r\u00e9duire la complexit\u00e9 associ\u00e9e \u00e0 la segmentation et \u00e0 la simplification 3D, nous avons projet\u00e9 les donn\u00e9es dans le domaine 2D pour effectuer la segmentation et avons ensuite cartographi\u00e9 les r\u00e9sultats dans le domaine 3D. Nous avons illustr\u00e9 ce cadre \u00e0 l’aide d’une fonction sp\u00e9cifique appel\u00e9e \u00ab\u00a0cha\u00eenes continues\u00a0\u00bb pour expliquer le processus de simplification. Par la suite, nous avons r\u00e9alis\u00e9 une analyse comparative par rapport \u00e0 des techniques de pointe \u00e9tablies, d\u00e9montrant la performance sup\u00e9rieure de notre m\u00e9thodologie. Dans nos futures recherches, nous visons \u00e0 \u00e9largir la port\u00e9e de notre cadre pour englober plusieurs caract\u00e9ristiques et les r\u00e9gions fonctionnelles \u00e0 l’int\u00e9rieur des mod\u00e8les CAO.<\/p>\n\n\n\n Abstract :<\/strong><\/p>\n\n\n\n In Industry 4.0 the use of Augmented Reality (AR) devices like HoloLens has gained significant acceptance for training assembly line operators in various industries. Simplifying 3D CAD models for AR training is crucial for better application performance. Our research focuses on developing methods and techniques to streamline complex 3D CAD models, making them suitable for AR applications. Our research highlights the role of 3D models in AR, enhancing the virtual experience by overlaying CAD models onto the real world. We detail AR applications in operator training and how integrating 3D CAD models improves understanding of instructions and procedures. We performed an extensive literature review in CAD model simplification for integration into augmented reality (AR) scenarios. Our findings indicated that mesh-based simplification techniques excel in preserving CAD model essentials, providing precise control over the level of details. Furthermore, we conducted four distinct types of assessments in our research. These assessments included objective evaluations utilizing mesh-based techniques from existing literature, expert review involving a comprehensive examination of each simplified model to determine the level of simplification based on vertex ranges, real-world testing aided by HoloLens2, which revealed improvements in framerate when using CAD models instead of their original versions. To conclude our evaluations, we also conducted user assessments, prioritizing user experience in our study. These assessments confirmed that the simplified models are highly capable of replacing the original counterparts. However, it was observed that further simplification is necessary, especially for intricate CAD models. The main methodology proposes an innovative approach, focusing on mesh segmentation and adaptive simplification using deep learning methods. To reduce the complexity associated with 3D segmentation and simplification we projected the data into the 2D domain to perform segmentation and mapped the results back to the 3D domain. We illustrated this framework using a specific feature called \u00ab\u00a0continuous chains\u00a0\u00bb to explain the simplification process. Subsequently, we conducted a comparative analysis against established state-of-the-art techniques, demonstrating the superior performance of our methodology. In our future research, we aim to broaden our framework’s scope to encompass multiple features and functional regions within CAD models.<\/p>\n","protected":false},"excerpt":{"rendered":" L’Ecole doctorale : Ecole Doctorale de l’Institut Polytechnique de Paris et le Laboratoire de recherche SAMOVAR – Services r\u00e9partis, Architectures, Mod\u00e9lisation, Validation, Administration des R\u00e9seaux pr\u00e9sentent l\u2019AVIS DE SOUTENANCE de Madame Abhaya Dhathri ARIGE Autoris\u00e9e \u00e0 pr\u00e9senter ses travaux en vue de l\u2019obtention du Doctorat de l’Institut Polytechnique de Paris, pr\u00e9par\u00e9 \u00e0 T\u00e9l\u00e9com SudParis en […]<\/p>\n","protected":false},"author":4,"featured_media":0,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"ocean_post_layout":"","ocean_both_sidebars_style":"","ocean_both_sidebars_content_width":0,"ocean_both_sidebars_sidebars_width":0,"ocean_sidebar":"0","ocean_second_sidebar":"0","ocean_disable_margins":"enable","ocean_add_body_class":"","ocean_shortcode_before_top_bar":"","ocean_shortcode_after_top_bar":"","ocean_shortcode_before_header":"","ocean_shortcode_after_header":"","ocean_has_shortcode":"","ocean_shortcode_after_title":"","ocean_shortcode_before_footer_widgets":"","ocean_shortcode_after_footer_widgets":"","ocean_shortcode_before_footer_bottom":"","ocean_shortcode_after_footer_bottom":"","ocean_display_top_bar":"default","ocean_display_header":"default","ocean_header_style":"","ocean_center_header_left_menu":"0","ocean_custom_header_template":"0","ocean_custom_logo":0,"ocean_custom_retina_logo":0,"ocean_custom_logo_max_width":0,"ocean_custom_logo_tablet_max_width":0,"ocean_custom_logo_mobile_max_width":0,"ocean_custom_logo_max_height":0,"ocean_custom_logo_tablet_max_height":0,"ocean_custom_logo_mobile_max_height":0,"ocean_header_custom_menu":"0","ocean_menu_typo_font_family":"0","ocean_menu_typo_font_subset":"","ocean_menu_typo_font_size":0,"ocean_menu_typo_font_size_tablet":0,"ocean_menu_typo_font_size_mobile":0,"ocean_menu_typo_font_size_unit":"px","ocean_menu_typo_font_weight":"","ocean_menu_typo_font_weight_tablet":"","ocean_menu_typo_font_weight_mobile":"","ocean_menu_typo_transform":"","ocean_menu_typo_transform_tablet":"","ocean_menu_typo_transform_mobile":"","ocean_menu_typo_line_height":0,"ocean_menu_typo_line_height_tablet":0,"ocean_menu_typo_line_height_mobile":0,"ocean_menu_typo_line_height_unit":"","ocean_menu_typo_spacing":0,"ocean_menu_typo_spacing_tablet":0,"ocean_menu_typo_spacing_mobile":0,"ocean_menu_typo_spacing_unit":"","ocean_menu_link_color":"","ocean_menu_link_color_hover":"","ocean_menu_link_color_active":"","ocean_menu_link_background":"","ocean_menu_link_hover_background":"","ocean_menu_link_active_background":"","ocean_menu_social_links_bg":"","ocean_menu_social_hover_links_bg":"","ocean_menu_social_links_color":"","ocean_menu_social_hover_links_color":"","ocean_disable_title":"default","ocean_disable_heading":"default","ocean_post_title":"","ocean_post_subheading":"","ocean_post_title_style":"","ocean_post_title_background_color":"","ocean_post_title_background":0,"ocean_post_title_bg_image_position":"","ocean_post_title_bg_image_attachment":"","ocean_post_title_bg_image_repeat":"","ocean_post_title_bg_image_size":"","ocean_post_title_height":0,"ocean_post_title_bg_overlay":0.5,"ocean_post_title_bg_overlay_color":"","ocean_disable_breadcrumbs":"default","ocean_breadcrumbs_color":"","ocean_breadcrumbs_separator_color":"","ocean_breadcrumbs_links_color":"","ocean_breadcrumbs_links_hover_color":"","ocean_display_footer_widgets":"default","ocean_display_footer_bottom":"default","ocean_custom_footer_template":"0","ocean_post_oembed":"","ocean_post_self_hosted_media":"","ocean_post_video_embed":"","ocean_link_format":"","ocean_link_format_target":"self","ocean_quote_format":"","ocean_quote_format_link":"post","ocean_gallery_link_images":"off","ocean_gallery_id":[],"footnotes":""},"categories":[286,169],"tags":[],"class_list":["post-6402","post","type-post","status-publish","format-standard","hentry","category-fractualites-ennews-fr","category-seminaires-armedia","entry"],"_links":{"self":[{"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/posts\/6402","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/users\/4"}],"replies":[{"embeddable":true,"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/comments?post=6402"}],"version-history":[{"count":1,"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/posts\/6402\/revisions"}],"predecessor-version":[{"id":6403,"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/posts\/6402\/revisions\/6403"}],"wp:attachment":[{"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/media?parent=6402"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/categories?post=6402"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/tags?post=6402"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}
M. Valeriu VRABIE<\/strong>, Professor, Universit\u00e9 de Reims Champagne-Ardenne, FRANCE – Rapporteur
M. Jean-Marc LECAILLEC<\/strong>, Professor, IMT Atlantique, FRANCE – Rapporteur
M. Marius PREDA<\/strong>, Ma\u00eetre de conf\u00e9rences, T\u00e9lecom SudParis, FRANCE – CoDirecteur de these
M. Francisco MOR\u00c1N BURGOS<\/strong>, Professeur, Universidad Polit\u00e9cnica de Madrid, ESPAGNE – Examinateur
Mme Arianne HINDS<\/strong>, Docteur, Tencent America, ETATS-UNIS – Examinateur<\/p>\n\n\n\n\u00ab Simplification des mod\u00e8les CAO 3D avec l’apprentissage profond pour la r\u00e9alit\u00e9 augment\u00e9e \u00bb<\/h2>\n\n\n\n
pr\u00e9sent\u00e9 par Madame Abhaya Dhathri ARIGE<\/h2>\n\n\n\n