Examinando por Autor "Otaduy, Miguel A."

Mostrando 1 - 20 de 23

A Bending Model for Nodal Discretizations of Yarn-Level Cloth
(Wiley, 2020) Pizana, José M.; Rodríguez, Alejandro; Cirio, Gabriel; Otaduy, Miguel A.
To deploy yarn-level cloth simulations in production environments, it is paramount to design very efficient implementations, which mitigate the cost of the extremely high resolution. To this end, nodal discretizations aligned with the regularity of the fabric structure provide an optimal setting for efficient GPU implementations. However, nodal discretizations complicate the design of robust and controllable bending. In this paper, we address this challenge, and propose a model of bending that is both robust and controllable, and employs only nodal degrees of freedom. We extract information of yarn and fabric orientation implicitly from the nodal degrees of freedom, with no need to augment the model explicitly. But most importantly, and unlike previous formulations that use implicit orientations, the computation of bending forces bears no overhead with respect to other nodal forces such as stretch. This is possible by tracking optimal orientations efficiently. We demonstrate the impact of our bending model in examples with controllable anisotropy, as well as ironing, wrinkling, and plasticity.
A computational optimization approach to scoliosis brace design – Preliminary results
(orthopaedic research society, 2022) Kardash, Kateryna; Rodriguez, Jesus P.; Otaduy, Miguel A.
The design of braces for adolescent idiopathic scoliosis is a problem that could potentially benefit from computational solutions. There is a broad body of research on simulation of the human spine, including simulation of scoliosis, as well as braces and their effects on the body. Some sparse works leverage these simulation models to evaluate the quality of digital brace designs, but the computational approach to scoliosis brace design remains largely underused.
A study of the sensitivity of biomechanical models of the spine for scoliosis brace design
(Elsevier, 2021) Koutras, Christos; Pérez, Jesús; Kardash, Kateryna; Otaduy, Miguel A.
Background and Objective: The development of biomechanical models of the torso and the spine opens the door to computational solutions for the design of braces for adolescent idiopathic scoliosis. However, the design of such biomechanical models faces several unknowns, such as the correct identification of relevant mechanical elements, or the required accuracy of model parameters. The objective of this study was to design a methodology for the identification of the aforementioned elements, with the purpose of creating personalized models suited for patient-specific brace design and the definition of parameter estimation criteria. Methods: We have developed a comprehensive model of the torso, including spine, ribcage and soft tissue, and we have developed computational tools for the analysis of the model parameters. With these tools, we perform an analysis of the model under typical loading conditions of scoliosis braces. Results: We present a complete sensitivity analysis of the models mechanical parameters and a comparison between a reference healthy subject and a subject suffering from scoliosis. Furthermore, we make a direct connection between error bounds on the deformation and tolerances for parameter estimation, which can guide the personalization of the model. Conclusions: Not surprisingly, the stiffness parameters that govern the lateral deformation of the spine in the frontal plane are some of the most relevant parameters, and require careful modeling. More surprisingly, their relevance is on par with the correct parameterization of the soft tissue of the torso. For scoliosis patients, but not for healthy subjects, we observe that the axial rotation of the spine also requires careful modeling.
Fast Numerical Coarsening with Local Factorizations
(Wiley, 2023) He, Zhongyun; Pérez, Jesús; Otaduy, Miguel A.
Numerical coarsening methods offer an attractive methodology for fast simulation of objects with high-resolution heterogeneity. However, they rely heavily on preprocessing, and are not suitable when objects undergo dynamic material or topology updates. We present methods that largely accelerate the two main processes of numerical coarsening, namely training data generation and the optimization of coarsening shape functions, and as a result we manage to leverage runtime numerical coarsening under local material updates. To accelerate the generation of training data, we propose a domain-decomposition solver based on substructuring that leverages local factorizations. To accelerate the computation of coarsening shape functions, we propose a decoupled optimization of smoothness and data fitting. We evaluate quantitatively the accuracy and performance of our proposed methods, and we show that they achieve accuracy comparable to the baseline, albeit with speed-ups of orders of magnitude. We also demonstrate our methods on example simulations with local material and topology updates.
Fine Virtual Manipulation with Hands of Different Sizes
(GMRV Publications, 2021) Sorli, Suzanne; Verschoor, Mickeal; Casas, Dan; Tajadura-Jiménez, Ana; Otaduy, Miguel A.
Natural interaction with virtual objects relies on two major technology components: hand tracking and hand-object physics simulation. There are functional solutions for these two components, but their hand representations may differ in size and skeletal morphology, hence making the connection non-trivial. In this paper, we introduce a pose retargeting strategy to connect the tracked and simulated hand representations, and we have formulated and solved this hand retargeting as an optimization problem. We have also carried out a user study that demonstrates the effectiveness of our approach to enable fine manipulations that are slow and awkward with na¨ıve approaches.
Learning Contact Corrections for Handle-Based Subspace Dynamics
(ACM, 2021) Casas, Dan; Pérez, Jesús; Otaduy, Miguel A.; Romero, Cristian
This paper introduces a novel subspace method for the simulation of dynamic deformations. The method augments existing linear handle-based subspace formulations with nonlinear learning-based corrections parameterized by the same subspace. Together, they produce a compact nonlinear model that combines the fast dynamics and overall contact-based interaction of subspace methods, with the highly detailed deformations of learning-based methods. We propose a formulation of the model with nonlinear corrections applied on the local undeformed setting, and decoupling internal and external contact-driven corrections. We define a simple mapping of these corrections to the global setting, an efficient implementation for dynamic simulation, and a training pipeline to generate examples that efficiently cover the interaction space. Altogether, the method achieves unprecedented combination of speed and contact-driven deformation detail.
Learning-Based Animation of Clothing for Virtual Try-On
(2020-04-17) Santesteban, Igor; Otaduy, Miguel A.; Casas, Dan
This paper presents a learning-based clothing animation method for highly efficient virtual try-on simulation. Given a garment, we preprocess a rich database of physically-based dressed character simulations, for multiple body shapes and animations. Then, using this database, we train a learning-based model of cloth drape and wrinkles, as a function of body shape and dynamics. We propose a model that separates global garment fit, due to body shape, from local garment wrinkles, due to both pose dynamics and body shape. We use a recurrent neural network to regress garment wrinkles, and we achieve highly plausible nonlinear effects, in contrast to the blending artifacts suffered by previous methods. At runtime, dynamic virtual try-on animations are produced in just a few milliseconds for garments with thousands of triangles. We show qualitative and quantitative analysis of results.
MDScale: Scalable multi-GPU bonded and short-range molecular dynamics
(Elsevier, 2021) Barreales, Gonzalo Nicolas; Novalbos, Marcos; Otaduy, Miguel A.; Sanchez, Alberto
GPUs have enabled a drastic change to computing environments, making massively parallel computing possible. Molecular dynamics is a perfect candidate problem for massively parallel computing, but to date it has not taken full advantage of multi-GPU environments due to the difficulty of partitioning molecular dynamics problems and exchanging problem data among compute nodes. These difficulties restrict the use of GPUs to only some of the computations in a full molecular dynamics problem, and hence prevent scalability beyond just a few GPUs. This work presents a scalable parallelization solution for the bonded and short-range forces present in a molecular dynamics problem. Together with existing solutions for long-range forces, it enables highly scalable, parallel molecular dynamics on multi-GPU computing environments. Specifically, the proposed solution divides the molecular volume into independent parts assigned to different GPUs, but it maintains a global bond structure that is efficiently exchanged when atoms move across GPUs. We demonstrate close-to-linear speedup of the proposed solution, simulating the dynamics of gigamolecules with 1 billion atoms on a computing environment with 96 GPUs, and obtaining superior performance to the well known molecular dynamics simulator NAMD.
Mixing Yarns and Triangles in Cloth Simulation
(2020-04-15) Casafranca, Juan J.; Cirio, Gabriel; Rodríguez, Alejandro; Miguel, Eder; Otaduy, Miguel A.
This paper presents a method to combine triangle and yarn models in cloth simulation, and hence leverage their best features. The majority of a garment uses a triangle-based model, which reduces the overall computational and memory cost. Key areas of the garment use a yarn-based model, which elicits rich effects such as structural nonlinearity and plasticity. To combine both models in a seamless and robust manner, we solve two major technical challenges. We propose an enriched kinematic representation that augments triangle-based deformations with yarn-level details. Naïve enrichment suffers from kinematic redundancy, but we devise an optimal kinematic filter that allows a smooth transition between triangle and yarn models. We also introduce a preconditioner that resolves the poor conditioning produced by the extremely different inertia of triangle and yarn nodes. This preconditioner deals effectively with rank deficiency introduced by the kinematic filter. We demonstrate that mixed yarns and triangles succeed to efficiently capture rich effects in garment fit and drape.
Modeling and Estimation of Nonlinear Skin Mechanics for Animated Avatars
(2020-04-17) Romero, Cristian; Otaduy, Miguel A.; Casas, Dan; Perez, Jesus
Data-driven models of human avatars have shown very accurate representations of static poses with soft-tissue deformations. However they are not yet capable of precisely representing very nonlinear deformations and highly dynamic effects. Nonlinear skin mechanics are essential for a realistic depiction of animated avatars interacting with the environment, but controlling physics-only solutions often results in a very complex parameterization task. In this work, we propose a hybrid model in which the soft-tissue deformation of animated avatars is built as a combination of a data-driven statistical model, which kinematically drives the animation, an FEM mechanical simulation. Our key contribution is the definition of deformation mechanics in a reference pose space by inverse skinning of the statistical model. This way, we retain as much as possible of the accurate static data-driven deformation and use a custom anisotropic nonlinear material to accurately represent skin dynamics. Model parameters including the heterogeneous distribution of skin thickness and material properties are automatically optimized from 4D captures of humans showing soft-tissue deformations.
Natural Tactile Interaction with Virtual Clay
(IEEE World Haptics Conference (WHC), 2021) Barreiro, Héctor; Torres, Joan; Otaduy, Miguel A.
Despite many past efforts to develop haptic experiences of virtual clay modeling, natural interaction with virtual clay remains a hard challenge. In this work, we propose a computational solution for the interactive simulation of clay-like materials with unprecedented realism, coupled with free-air tactile rendering that provides a natural tangible experience. Our solution includes a novel particle-based model of viscoplasticity for efficient interactive simulation, and an optimization-based ultrasound rendering algorithm that takes as input the interaction forces between a virtual hand model and the clay-like material. We demonstrate the effectiveness of our method through expressive creative experiences.
Non-invasive procedure for acquisition of mechanical properties of the torso
(-, 2022) Koutras, Christos; Shayestehpour, Hamed; Perez, Jesus; Wong, Christian; Arnesen, Anna; Rasmussen, John; Otaduy, Miguel A.
Computational methods promise benefits for the design of braces to manage adolescent idiopathic scoliosis. However, computational methods for the design of scoliosis braces suffer an important challenge: they require a personalized model of the patient’s torso biomechanics. The biggest difficulty in building a personalized model of the torso is defining its mechanical parametrization. In this work, we present a non-invasive procedure to obtain simultaneously force and deformation that characterize the mechanical response of the torso. We have tested the method on ten scoliotic patients, and we demonstrate its sensitivity by quantifying the range of forces and Cobb angles during the procedure.
Parametric Skeletons with Reduced Soft-Tissue Deformations
(Wiley, 2021) Tapia, Javier; Romero, Cristian; Pérez, Jesús; Otaduy, Miguel A.
We present a method to augment parametric skeletal models with subspace soft-tissue deformations. We combine the benefits of data-driven skeletal models, i.e., accurate replication of contact-free static deformations, with the benefits of pure physicsbased models, i.e., skin and skeletal reaction to contact and inertial motion with two-way coupling. We succeed to do so in a highly efficient manner, thanks to a careful choice of reduced model for the subspace deformation. With our method, it is easy to design expressive reduced models with efficient yet accurate force computations, without the need for training deformation examples. We demonstrate the application of our method to parametric models of human bodies, SMPL, and hands, MANO, with interactive simulations of contact with nonlinear soft-tissue deformation and skeletal response
Path Routing Optimization for STM Ultrasound Rendering
(Institute of Electrical and Electronics Engineers, 2020-02) Barreiro, Hector; Sinclair, Stephen; Otaduy, Miguel A.
Ultrasound transducer arrays are capable of producing tactile sensations on the hand, promising hands-free haptic interaction for virtual environments. However, controlling such an array with respect to reproducing a desired perceived interaction remains a challenging problem. In this work we approach this problem as a dynamic mapping of virtual interactions to existing control metaphors of ultrasound devices, namely, the modulation of focal point positions and intensities over time, a method known as Spatiotemporal Modulation (STM). In particular, we propose an optimization approach that takes into account known perceptual parameters and limitations of the STM method. This results in a set of focal point paths optimized to best reconstruct an arbitrary target pressure field.
Real-time Pose and Shape Reconstruction of Two Interacting Hands With a Single Depth Camera
(ACM Transactions on Graphics, 2019) Mueller, Franziska; Davis, Micah; Bernard, Florian; Sotnychenko, Oleksandr; Verschoor, Mickeal; Otaduy, Miguel A.; Casas, Dan; Theobalt, Christian
Wepresentanovelmethodforreal-timeposeandshapereconstructionof twostronglyinteractinghands.Ourapproachisthefirsttwo-handtracking solutionthatcombinesanextensivelistoffavorableproperties,namelyitis marker-less,usesasingleconsumer-leveldepthcamera,runsinrealtime, handlesinter-andintra-handcollisions,andautomaticallyadjuststothe user’shandshape.Inordertoachievethis,weembedarecentparametric handposeandshapemodelandadensecorrespondencepredictorbasedon adeepneuralnetworkintoasuitableenergyminimizationframework.For trainingthecorrespondencepredictionnetwork,wesynthesizeatwo-hand dataset based on physical simulations that includes both hand pose and shapeannotationswhileatthesametimeavoidinginter-handpenetrations. Toachievereal-timerates,wephrasethemodelfittingintermsofanonlinear least-squaresproblemsothattheenergycanbeoptimizedbasedonahighly efficient GPU-based Gauss-Newton optimizer. We show state-of-the-art resultsinscenesthatexceedthecomplexityleveldemonstratedbyprevious
Rendering of Constraints with Underactuated Haptic Devices
(Institute of Electrical and Electronics Engineers, 2020-03-20) Lobo, Daniel; Otaduy, Miguel A.
Several previous works have studied the application of proxy-based rendering algorithms to underactuated haptic devices. However, all these works make oversimplifying assumptions about the configuration of the haptic device, and they ignore the user’s intent. In this work, we lift those assumptions, and we carry out a theoretical study that unveils the existence of unnatural ghost forces under typical proxy-based rendering. We characterize and quantify those ghost forces. In addition, we design a novel rendering strategy, with anisotropic coupling between the device and the proxy. With this strategy, the forces rendered by an underactuated device are a best match of the forces rendered by a fully actuated device. We have demonstrated our findings on synthetic experiments and a simple real-world experiment.
Robust Eulerian-on-Lagrangian Rods
(Association for Computing Machinery (ACM), 2020) Sánchez-Banderas, Rosa M.; Rodríguez, Alejandro; Barreiro, Héctor; Otaduy, Miguel A.
This paper introduces a method to simulate complex rod assemblies and stacked layers with implicit contact handling, through Eulerian-on-Lagrangian (EoL) discretizations. Previous EoL methods fail to handle such complex situations, due to ubiquitous and intrinsic degeneracies in the contact geometry, which prevent the use of remeshing and make simulations unstable. We propose a novel mixed Eulerian-Lagrangian discretization that supports accurate and efficient contact as in EoL methods, but is transparent to internal rod forces, and hence insensitive to degeneracies. By combining the standard and novel EoL discretizations as appropriate, we derive mixed statics-dynamics equations of motion that can be solved in a unified manner with standard solvers. Our solution is simple and elegant in practice, and produces robust simulations on large-scale scenarios with complex rod arrangements and pervasive degeneracies. We demonstrate our method on multi-layer yarn-level cloth simulations, with implicit handling of both intraand inter-layer contacts.
Self-Supervised Collision Handling via Generative 3D Garment Models for Virtual Try-On
(GMRV Publications, 2021) Santesteban, Igor; Thuerey, Nils; Otaduy, Miguel A.; Casas, Dan
We propose a new generative model for 3D garment deformations that enables us to learn, for the first time, a data-driven method for virtual try-on that effectively addresses garment-body collisions. In contrast to existing methods that require an undesirable postprocessing step to fix garment-body interpenetrations at test time, our approach directly outputs 3D garment configurations that do not collide with the underlying body. Key to our success is a new canonical space for garments that removes pose-and-shape deformations already captured by a new diffused human body model, which extrapolates body surface properties such as skinning weights and blendshapes to any 3D point. We leverage this representation to train a generative model with a novel self-supervised collision term that learns to reliably solve garment-body interpenetrations. We extensively evaluate and compare our results with recently proposed data-driven methods, and show that our method is the first to successfully address garment-body contact in unseen body shapes and motions, without compromising realism and detail.
Simulation-Based Morphing of Personalized Models of the Torso for Scoliosis Brace Design – Preliminary Results
(orthopaedic research society, 2022) Koutras, Christos; Shayestehpour, Hamed; Rodriguez, Jesus P.; Rasmussen, John; Wong, Christian; Otaduy, Miguel A.
Computational and personalized design of braces for patients suffering from adolescent idiopathic scoliosis is a subject that has not been extensively studied and faces several unknowns. One of the most challenging tasks is the development of patient-specific biomechanical models of the torso. The first step required in order to build a personalized model is the acquisition of the patient’s specific geometry, i.e., the bones and joints of the skeleton. To this end, this study morphs a template torso model into patient-specific data in the following way: We start by acquiring x-rays, and we annotate personalized landmarks. Then, we use as template a biomechanical torso model consisted of a multibody dynamic system coupled with FEM and we proceed to the morphing in two steps. First, an initial tuning by adjusting the global scale of the model and finally a finer one, taking into account local deformations.
Soft-Tissue Simulation for Computational Planning of Orthognathic Surgery
(MDPI, 2021) Alcañiz, Patricia; Pérez, Jesús; Gutiérrez, Alessandro; Barreiro, Héctor; Villalobos, Ángel; Miraut, David; Illana, Carlos; Guiñales, Jorge; Otaduy, Miguel A.
Simulation technologies offer interesting opportunities for computer planning of orthognathic surgery. However, the methods used to date require tedious set up of simulation meshes based on patient imaging data, and they rely on complex simulation models that require long computations. In this work, we propose a modeling and simulation methodology that addresses model set up and runtime simulation in a holistic manner. We pay special attention to modeling the coupling of rigid-bone and soft-tissue components of the facial model, such that the resulting model is computationally simple yet accurate. The proposed simulation methodology has been evaluated on a cohort of 10 patients of orthognathic surgery, comparing quantitatively simulation results to post-operative scans. The results suggest that the proposed simulation methods admit the use of coarse simulation meshes, with planning computation times of less than 10 seconds in most cases, and with clinically viable accuracy.