Archive ouverte de Centrale Lyon
Not a member yet
32420 research outputs found
Sort by
Reimagining Urban River Bathing in Europe: A Multisectoral and Interdisciplinary Dive Into Lyon's Rivers (France)
No full textInternational audienceUrban river bathing is re‐emerging across Europe, driven by social demand and climate change impacts. The Urban Bathing Consortium, an interdisciplinary and intersectoral consortium initiated at the University of Lyon (France), is at the forefront of studying the challenges and opportunities of creating and managing healthy, safe, and accessible river bathing spaces. Through interdisciplinary collaboration among researchers and stakeholders, the consortium proposed an analytical framework, identifying seven critical dimensions for urban river bathing: the history and revival of city‐river relationships, legal and regulatory frameworks, bathing water quality, river drowning risks, river ecosystems, social perspectives, and urban planning. By examining these dimensions with state‐of‐the‐art approaches and drawing on Lyon's experiences, the study provides scientific insights and practical recommendations for future sustainable urban river bathing development. These include revitalizing historical city‐river connections, aligning local regulations with EU guidance, advancing holistic microbial water quality control, enhancing safety measures, incorporating ecological considerations, balancing competing river uses in urban planning, and addressing social needs for inclusive river governance
Spreading properties of the Fisher--KPP equation when the intrinsic growth rate is maximal in a moving patch of bounded size
No full textInternational audienceThis paper is concerned with spreading properties of space-time heterogeneous Fisher–KPP equations in one space dimension. We focus on the case of everywhere favorable environment with three different zones, a left half-line with slow or intermediate growth, a central patch with fast growth and a right half-line with slow or intermediate growth. The central patch moves at various speeds. The behavior of the front changes drastically depending on the speed of the central patch. Among other things, intriguing phenomena such as nonlocal pulling and locking may occur, which would make the behavior of the front further complicated. The problem we discuss here is closely related to questions in biomathematical modelling. By considering several special cases, we illustrate the remarkable diversity of possible behaviors. In particular, when the central patch has constant size and constant speed, we provide a complete set of explicit formulas for the spreading speed
Jacob Kiplimo's Barcelona Half-Marathon Record: What Is the True Impact of Running Behind the Lead Vehicle?
No full textDeploying Compact and Reliable Deep Neural Networks in Safety-critical Applications
No full textInternational audienc
When Curvature Counts: Hyperbolic Geometry in Prototype-Based Image Classification
No full textInternational audiencePrototype Learning offers an interpretable and efficient classification framework by mapping data into an embedding space structured around class prototypes. Recent research has explored non-Euclidean geometries, such as hyperspherical and hyperbolic spaces, to more effectively model latent hierarchical structures and complex data relationships. While these geometries have shown potential, leveraging them within an image classification context is not trivial. To address this, we propose HypPNet, a hyperbolic prototypical model on the Poincaré ball that integrates Riemannian optimization and norm-based regularization to perform effectively without prior data knowledge. Experiments on three benchmark datasets and multiple embedding dimensions show that HypPNet outperforms its competitors across alternative geometries, improving classification performance over various metrics
For the use of exterior form in daily physics, an introduction without coordinate frame
No full textThis is a short introduction of the exterior form formalism focus on its appli-cations in physics and then mostly aimed to physics students. If exterior formsare more than a century old they are unfortunately still seen (and teached) asa high level mathematics object and then little used outside theorical physics.We then focus here on simple examples which occure in daily phiysics. Thereexists already a lot of very good mathematical textbooks and courses on thesubject but the originality of these notes, the physical applications aside, is thatwe keep a completely geometrical approach. As a rule of a game played here wenever use a coordinate frame neither in the definitions nor in the proofs but onlyat the end in order to recover the classical physics equations
Dual-arm motion-compensated single-pixel imaging
No full textSingle-pixel imaging offers a cost-effective strategy for high-resolution imaging over a wide range of electromagnetic frequencies, but suffers from motion artifacts due to its sequential acquisition process. In this work, we propose a new framework for dynamic single-pixel imaging that is particularly suited for hyperspectral imaging. First, we introduce a hybrid dual-arm device combining a hyperspectral single-pixel camera and a conventional imaging arm, which allows accurate motion estimation during acquisition. We then reformulate the reconstruction problem by compensating for motion, thereby reducing the dynamic problem to a static reconstruction task over an extended field of view. Two different discretizations -warping either the illumination patterns or the image to be reconstructed are proposed, along with an in-depth analysis of their tradeoffs. Through extensive numerical simulations and real-world experiments, we demonstrate that warping the image rather than the patterns leads to superior reconstruction quality. In addition, extending the field of view beyond that of the single-pixel camera significantly mitigates model mismatch and improves image fidelity. The proposed method achieves low computational cost while maintaining theoretical rigor, providing a practical and robust solution for dynamic single-pixel imaging. Open-source implementations are provided to facilitate reproducibility and future research.</div
Convergence analysis of semi-smooth Newton method for mixed FEM approximations of dynamic two-body contact and crack problems
No full textInternational audienceA class of elastodynamic problems describing contact between two deformable bodies as well as non-penetrating cracks in a single body is considered in the framework of FEM approximation. For time discretization, the Hilber-Hughes-Taylor (HHT-alpha) method extending Newmark schemes is incorporated. Using mixed variational formulation of the fully discrete contact problem, a semi-smooth Newton method of solution is provided with the locally super-linear convergence. An equivalent primal-dual active set algorithm validates monotone properties of global convergence for the Newton iterates provided by M-matrix property. Numerical solution of the Signorini contact with rigid obstacle is presented for isotropic body in 2D using benchmark and moving load experiment
Survey on differential estimators for 3d point clouds
No full textInternational audienceRecent advancements in 3D scanning technologies, including LiDAR and photogrammetry, have enabled the precise digital replication of real-world objects. These methods are widely used in fields such as GIS, robotics, and cultural heritage. However, the point clouds generated by such scans are often noisy and unstructured, posing challenges for traditional geometry processing tasks. Accurately estimating differential properties like surface curvatures and normals is crucial for tasks such as shape matching and classification, but remains complex due to these inherent challenges. This paper reviews state-of-the-art methods for estimating differential properties from 3D point clouds, with a focus on approaches that offer strong mathematical foundations and theoretical guarantees.We also benchmark these methods using various datasets, evaluating their performance in terms of accuracy, robustness, and efficiency. Our contributions include the release of datasets, tools, and code to promote reproducibility and support future research in this area