277989 research outputs found
Sort by
Towards Gestural Interaction with Cubic Tangible Objects in Learning Activities
No full textCubic tangible interfaces are a promising yet underexplored modality in educational settings that could inform their development. This paper investigates gestural interaction with cubic objects, specifically AudioCubes, in learning contexts. Using the Goal-Questions-Metrics framework as a foundation, this study employs a systematic literature review, gesture elicitation studies, and usability evaluations to determine how cubic tangible interfaces can be adapted to meet the needs of diverse users in various educational contexts. It examines how cubic tangible interfaces can support learning activities by identifying preferred gestures and movement patterns across various tasks and user contexts. This highlights the need for a structured engineering methodology that supports the creation of reusable, generalizable cubic tangible interface designs, moving away from one-off designs that are often valid only for a particular case. The study classifies core gesture types using µGlyph notation to enable consistent representation across tasks. Usability assessments reveal common issues, user preferences, and ways to enhance user immersion. Furthermore, the research emphasizes adapting interfaces for diverse users, especially younger learners and individuals with different motor or cognitive abilities
P2 purinergic receptors: A key but underexplored player in post-myocardial infarction healing.
No full textEllipsoidal Embeddings of Graphs
No full textDue to their flexibility to represent almost any kind of relational data, graph-based models have enjoyed tremendous success over the past decades. While graphs are inherently only combinatorial objects, however, many prominent analysis tools are based on the algebraic representation of graphs via matrices such as the graph Laplacian, or on associated graph embeddings. Such embeddings associate to each node a set of coordinates in a vector space, a representation that can then be employed for learning tasks such as the classification or alignment of the nodes of the graph. As the geometric picture provided by embedding methods enables the use of a multitude of methods developed for vector space data, embeddings have thus gained interest from a theoretical as well as a practical perspective. Inspired by trace optimization problems, often encountered in the analysis of graph-based data, here we present a method to derive ellipsoidal embeddings of the nodes of a graph, in which each node is assigned a set of coordinates in a hyperellipsoid. Our method may be seen as an alternative to popular spectral embedding techniques, with which it shares certain similarities we discuss. To illustrate the utility of the embedding we conduct a case study in which we analyze synthetic and real world networks with modular structure, and compare the results obtained with known methods in the literature
Reversal of cerebrovascular anomalies in a zebrafish model of vein of Galen aneurysm
No full textCongenital vascular malformations result from abnormal development of the vascular tree, with the aneurysmal malformation of the vein of Galen (VGAM) being the most prevalent neurovascular malformation in neonates, associated with poor outcomes. This condition is linked to germline mutations in the RASA1 and EPHB4 genes, although the underlying developmental mechanisms remain unclear. Here we generate zebrafish models lacking rasa1a and ephb4a that replicate the genetic and structural features of VGAMs. Our findings connect the development of malformations to insufficient fusion of precursor blood vessels, a process regulated by blood flow and the responses of endothelial cells. RASA1 deficiency destabilizes the homeostatic response to blood flow and contributes to impaired flow-mediated activation of MAPK and phosphatidylinositol-3-kinase signaling. By pharmacologically targeting these signaling pathways in mutant models, we restore normal fusion in existing malformations, offering potential new strategies for treating VGAMs and similar vascular remodeling disorders
Person Re-Identification and its application to Multi-Object Tracking
No full textMulti-Object Tracking (MOT) is a fundamental computer vision task that involves detecting objects of interest in video frames and associating detections of the same object across time to form trajectories. For association, appearance cues, extracted through person re-identification (ReID) models, play a crucial role by capturing distinctive visual features of the tracked targets. However, despite its importance for tracking, ReID has primarily been studied as an image retrieval problem, with state-of-the-art methods overlooking tracking-specific challenges. This thesis focuses on advancing person re-identification methods, with a particular emphasis on making them more robust and better suited for tracking applications. A key challenge in ReID and MOT is handling occlusions, where targets become partially hidden by objects or other people, leading to degraded re-identification accuracy and potential identity switches in tracking. Additionally, tracking methods often fail to effectively combine ReID with motion cues and scene context, relying instead on naive association strategies. To address these challenges, this thesis makes three key contributions: BPBreID, a part-based method for robust occluded re-identification; KPR, a keypoint promptable ReID model designed to address multi-person occlusions scenarios; and CAMELTrack, an online tracking-by-detection method that replaces traditional heuristic for detection association with a context-aware learnable module. At the time of writing, KPR and CAMELTrack achieve state-of-the-art performance on widely-used benchmarks for occluded person re-identification and multi-object tracking. Complementary to this thesis work, additional research contributions were made in sports analytics, including the development of specialized re-identification models for athletes and the introduction of novel datasets for player tracking, re-identification, and jersey number recognition. The thesis concludes with a critical analysis of the current state of tracking and re-identification technologies, offering my opinionated view about the future of these two rapidly evolving fields.(FSA - Sciences de l'ingénieur) -- UCL, 202
Construire l’intersubjectivité à distance : étude des effets de l’explicitation des intentions techniques et pédagogiques sur l’apprentissage avec une vidéo interactive
No full textAs higher education increasingly embraces digital formats, interactive video stands out as a tool for supporting learning, especially in remote and asynchronous contexts. Yet, the spatial and temporal separation it entails shifts the responsibility for communication onto the design and the learner's ability to interpret it. Given the constraints of human-computer interaction, this often limits mutual understanding—particularly among students facing digital inequalities or struggling to engage with interactive features. This study investigates whether making the instructor’s pedagogical and technical intentions explicit can help mitigate the effects of their physical absence and improve learning outcomes. While pedagogical intentions aim to guide learners in their learning process, technical intentions concern opportunities for interaction with the tool. The underlying hypothesis is that a prior understanding of both dimensions supports better engagement with content and the achievement of educational goals. To test this, a mixed-method approach was adopted, combining a controlled experiment with qualitative data. Findings reveal that individual learner characteristics—such as socioeconomic background, digital experience and prior knowledge—significantly shape how students interact with interactive video, often diminishing the intended benefits of explicit guidance. These insights highlight the need to further investigate the social dynamics at play in digital learning environments in order to design more inclusive and effective educational tools.Dans un contexte de numérisation croissante de l’enseignement supérieur, la vidéo interactive émerge comme un dispositif numérique propice à l’enrichissement des processus éducatifs, notamment dans les situations distancielles et asynchrones. Toutefois, la rupture spatio-temporelle inhérente à ce mode d’enseignement fait reposer la communication sur le design pédagogique élaboré par l’enseignant ainsi que sur la capacité de l’apprenant à en saisir les intentions implicites. Encadrée par les contraintes de l’interaction humain-machine, cette forme de communication réduit les opportunités d’intercompréhension alors même que des disparités socionumériques et des difficultés d’appropriation des fonctionnalités interactives sont constatées au sein de la population étudiante. Face à ces enjeux, cette recherche s’est attachée à interroger l’utilité d’expliciter les intentions pédagogiques et techniques de l’enseignant de manière à compenser son absence physique dans la relation d’apprentissage. Les premières visent à guider l’apprenant dans la gestion de son apprentissage, tandis que les secondes renvoient aux opportunités d’interaction avec le dispositif. L’hypothèse posée est que la compréhension préalable du fonctionnement de l’outil et de ses implications pour l’apprentissage facilite l’assimilation du matériau didactique et favorise l’atteinte des objectifs éducatifs. Afin d’évaluer les effets de l’explicitation sur les performances d’apprentissage, une approche méthodologique mixte a été déployée, articulant une expérimentation contrôlée et une collecte de données qualitatives. Les résultats mettent en évidence l’influence significative des caractéristiques individuelles des apprenants, telles que l’origine socioéconomique, l’expérience numérique et les connaissances préalables, sur les modalités d’interaction avec le dispositif interactif, réduisant ainsi l’efficacité de l’explicitation. Ce constat invite à poursuivre l’examen des dynamiques sociales à l’œuvre dans les environnements numériques d’apprentissage dans le but de développer des dispositifs performants et inclusifs.(COMU - Information et communication) -- UCL, 202
The dark side of the moon : l'expression des appartenances des Thraces dans l'Empire romain (Ier - IIIe s.)
No full textWhat becomes of the affiliations of individuals who have settled—willingly or not—in foreign lands? This question forms the backbone of this dissertation, which is devoted to the study of how 983 Thracians, settled or stationed in the Roman Empire during the first three centuries of our era, expressed their affiliations. By approaching affiliation as an objectifiable subject of study, we examine the socio-cultural and territorial markers of identity as they are revealed through 627 inscriptions. The significant mobility of the Thracians, closely tied to their widespread recruitment into various units of the Roman army, makes it possible to construct six case studies corresponding to distinct territorial areas (Rome, the Italian Peninsula, Roman Syria, the western and Danubian provinces, and North Africa). Each of these cases is systematically analyzed using a methodology structured around five perspectives : onomastics, origo (origin), sociability, religious practices, and iconographic representations. Although the results are specific to each case, they reveal potential disparities or similarities that transcend the isolated nature of each study. Ultimately, these insights contribute to a broader reflection on the very act of expressing affiliation in ancient societies. In this regard, what does the epigraphic expression of affiliation reveal about its users? To what extent do the different social environments with which the Thracians interact influence how they express their affiliations? Since affiliations are multiple and can coexist simultaneously, how are they articulated with one another? And finally, what drives individuals to reveal certain affiliations while concealing others?(HIAR - Histoire, art et archéologie) -- UCL, 202
The role of cultural values in the effectiveness of environmental NGO campaigns on sustainable meat consumption
No full textFood production, particularly meat consumption, poses severe environmental challenges, including greenhouse gas emissions, deforestation, water depletion, and nutrient runoff. Despite the urgency of reducing meat consumption, traditional norms and cultural values reinforce its centrality in Western diets, limiting the effectiveness of environmental campaigns. This study examines the underexplored role of individual cultural values in shaping responses to environmental NGO campaigns promoting sustainable meat consumption. Analyzing data from 514 respondents across five European countries with diverse national cultures, the research highlights the significant influence of three cultural values: power distance, masculinity, and uncertainty avoidance. These values were found to diminish the effectiveness of campaigns emphasizing environmental degradation and animal welfare. The findings contribute to the literature on sustainable food behaviors by underscoring the unique cultural dynamics involved in meat consumption at the individual level. Practical recommendations are offered to NGOs, policymakers, and managers for tailoring campaigns to address cultural nuances and drive sustainable dietary transitions
Climate change expected to increase conflict risks over the next decades across sub-Saharan Africa
No full textWhile climate change is increasingly recognized as a driver of conflict risks, most research focuses on past correlations, hence limiting our ability to project future climate-related security risks. We use machine learning techniques to predict the spatiotemporal dynamics of conflict risks and estimate the human population that would be exposed to these risks in the 2030s, 2040s, and 2050s. Our results demonstrate that climate change is associated with increasing risks of conflict over the next decades and that the change of projected spatial and temporal distribution of conflict risk will be heterogeneous, depending on conflict types and climate scenarios. We estimate that 0.5-1.7 billion people may live at high-risk zones of conflict across sub-Saharan Africa by the 2050s, which is an increase of at least 391 million compared to today. Detecting such risks early provides sufficient time to plan ahead and implement mitigation strategies