HAL Portal UPHF (Université Polytechnique Hauts-de-France)
Not a member yet
28473 research outputs found
Sort by
Poétiques de la musicologie
No full textThis dissertation looks back on a career that began with a fork in the road from philosophy to musicology. Attention to intertextuality between poetry and music opened up to questions of the interference of reception effects on poetic forms and, through radio experiments that became increasingly reflexive as they asserted themselves as musical mediations through creation, ended up opening up a path of research between intertextualities, intermedialities and interferences between poetry, music and mediations.Le mémoire de synthèse revient sur un parcours qui s’ouvre par une bifurcation de la philosophie vers la musicologie. L’attention aux intertextualité entre poésie et musique s’est ouverte aux questions des interférences des effets de réception sur les formes poétiques et, par des expérimentations radiophoniques de plus en plus réflexives à mesure qu’elles se sont revendiquées comme des médiations musicales par la création, a fini par ouvrir un chemin de recherche entre intertextualités, intermédialités et interférences entre poésie, musique et médiations
Investigation of the Effect of the Nitrogen on the Band Offset and the Intersubband Absorption Coefficient of the GaNxAs1-x-ySby/GaSb Quantum Well Structures
No full textInternational audienc
Miniaturized Intelligent Actuators Enabling Active Endoscopic Navigation
No full textInternational audienceThis poster presents the development of intelligent actuators for active micro-endoscopes, designed to enhance surgeons’ diagnostic capabilities during operations. The proposed system enables real-time adjustment of the viewing angle without physically moving the endoscope, thus contributing to less invasive procedures and minimizing the need for larger medical tools. The research addresses the challenge of creating a thin, flexible, and actively controlled endoscope by exploring the integration of the smallest commercially available image sensors and micro-actuators. Special attention is given to the miniaturization constraints and the biocompatibility of components in the targeted surgical environment.Ce poster présente le développement d'actionneurs intelligents pour micro-endoscopes actifs, conçus pour améliorer les capacités de diagnostic des chirurgiens pendant l’opération. Le système proposé permet d’ajuster l’angle de vue en temps réel sans déplacer physiquement l’endoscope, favorisant ainsi des interventions moins intrusives et limitant l’utilisation d’outils médicaux encombrants. La recherche traite du défi de concevoir un endoscope actif, fin et flexible, en intégrant les plus petits capteurs d’image et actionneurs disponibles sur le marché. Une attention particulière est portée aux contraintes de miniaturisation ainsi qu’à la biocompatibilité des composants dans l’environnement chirurgical ciblé
Implementing Real-Time Markerless Motion Capture in Smartphone Exergames
No full textHighlights:• Smartphone cameras using AI-powered pose estimation enable affordable and accessible exergames for rehabilitation and physical activity.• Balancing accuracy and latency is crucial, lightweight AI models improve responsiveness but may reduce movement precision.• User positioning affects visibility and tracking: solutions include larger UI elements, auditory feedback and upper-body-focused exercises.• Optimization of AI models and user studies are needed to improve usability in mobile exergames
Étude par microscopies à sondes locales de surfaces GaN gravées à l'argon pour la réalisation de diodes Schottky
No full textInternational audienc
« À quel verbiage vous amusez-vous depuis si longtemps Socrate ? ». Le dialogue antique comme instrument pédagogique et ses héritages contemporains.
No full textNational audienceWhen words want to speak, they are listened and they are read. Oral and writing are linked by the "système-langue" of the ancient Greeks. Interaction as a pedagogical process is questionned as far as ancient sources are concerned : ancient dialogues reveal didactical approches of orality through writing.Three goals must be reached thanks to this communication : the first one is to identify, caracterise and define what "orality" is in the sources. The second is to understand the different uses of oral, as a pedagogical tool, in ancient times. The third one is to emphasise the link with our own pedagogical contemporary approches, by interrogating oral in officials educationnal documents.Quand les mots veulent parler, ils s’écoutent et se lisent. Oral et écrit sont liés au prisme du système-langue. L'interaction d'un point de vue pédagogique est ici abordée du point de vue des auteurs antiques : les dialogues antiques sont des écrits révélateurs de pratiques didactiques de l'oral.Par conséquent, trois objectifs sont à atteindre dans cette conférence de séminaire : le premier consiste à identifier les différents concepts liés à la parole, à les caractériser et à en définir les clés d’analyse. Le deuxième est, à partir des termes choisis, d’appréhender les approches pédagogiques liées à l’oral dans les dialogues antiques. Le troisième est d’établir le lien avec l’époque contemporaine en questionnant les approches éducatives de l’oral dans les programmes scolaires du temps présen
Class-Specific Dataset Splitting for YOLOv8: Improving Real-Time Performance in NVIDIA Jetson Nano for Faster Autonomous Forklifts
No full textInternational audienceThis research examines a class-specific YOLOv8 model setup for real-time object detection using the Logistics Objects in Context dataset, specifically looking at how it can be used in high-speed autonomous forklifts to enhance obstacle detection. The dataset contains five common object classes in logistics warehouses. It is divided into transporting tools (forklift and pallet truck) and goods-carrying tools (pallet, small load carrier, and stillage) to meet specific task needs. Two YOLOv8 models were individually trained and implemented on the NVIDIA Jetson Nano, with each one specifically optimized for a tool category. Using this approach tailored to specific classes resulted in a 30.6 percent decrease in inference time compared to training a single YOLOv8 model on all classes. Task-specific detection saw a 74.4 percent improvement in inference time for transporting tools and 56.2 percent improvement for goods-carrying tools. Furthermore, the technique decreased the hypothetical distance traveled during inference from 45.14 cm to 31.32 cm and even as low as 11.55 cm for transporting tools detecting while still preserving detection accuracy with a minor drop of 1.25% in mean average precision. The integration of these models onto the NVIDIA Jetson Nano made this approach compatible for future autonomous forklifts and showcases the potential of the technique to improve industrial automation. This study demonstrates a useful and effective method for real-time object detection in intricate warehouse settings by matching detection tasks with practical needs