Figure 7 - uploaded by Jos Stam
Content may be subject to copyright.
Every velocity field is the sum of an incompressible field and a gradient field (top). To obtain an incompressible field we simply subtract the gradient field from our current velocities (bottom).
Source publication
In this paper we present a simple and rapid implementation of a fluid dynamics solver for game engines. Our tools can greatly enhance games by providing realistic fluid-like effects such as swirling smoke past a moving character. The potential applications are endless. Our algorithms are based on the physical equations of fluid flow, namely the Nav...
Similar publications
New minimum intact stability criteria are presented to ensure safety against capsizing invoked by sudden loss of crane load during heavy lifts at sea, followed by typical sample stability assessments for a lifting operation on four multipurpose ships. For added stability, two of these ships had a pontoon attached at their sides opposite the lift. T...
Mechanical analysis of movement plays an important role in clinical management of neurological and orthopedic conditions. There has been increasing interest in performing movement analysis in real-time, to provide immediate feedback to both therapist and patient. However, such work to date has been limited to single-joint kinematics and kinetics. H...
Citations
... A realistic and stable real-time representation of fluid has been a long-standing research issue in the field of Computer Graphics [1][2][3][4]. MC (Marching Cubes), a representative method for representing the surface of fluids, has the disadvantage of losing the model's sharp features [5]. DC (Dual Contouring), which overcomes this drawback, can accurately represent sharp features of objects using QEF (Quadratic Error Function), but surface normalization must precede to calculate the appropriate vertex positions [6]. ...
In this paper, we propose a learning model for tracking the isolines of fluid based on the physical properties of particles in particle-based fluid simulations. Our method involves analyzing which weights, closely related to surface tracking among the various physical properties of fluid particles, are significant. These weights are used as input values for the learning algorithm, enabling relatively accurate isoline tracking. In addition, compared to existing learning models such as linear regression, LSTM (long short-term memory), and learning representation (1-layer) models, our method obtained superior surface tracking results without accumulating errors. By using our proposed network structure to track the fluid surface, it learns and predicts values derived from existing fluid simulation algorithms, eliminating the need for computational processes for level-set values and enabling real-time surface tracking. As the scale of the simulation increases, our method significantly reduces the time and resources consumed compared to traditional methods and can track the fluid surface without additional resource consumption. Furthermore, due to our method’s simple network structure, the time consumed in the initial process of loading the model into memory is faster than models such as CNN and LSTM. Our proposed model occupies less than 30 kb of storage space, making it suitable for use in middleware. Lastly, to verify the generality of our method, we conducted tests in a total of five scenes, and in all test scenes, visually natural fluid isolines were represented.
... Mathias Berger and Verina Cristie [4] proposed using game engine technology to bridge the gap between architects and engineers in evaluating the effect of buildings on urban climate through CFD methods. Jos Stam [5] presented a rapid implementation of a fluid dynamics solver for game engines based on the physical equations of fluid flow, emphasizing stability and speed for just-in-time performance. Wangda Zuo and Qingyan Chen [6] proposed the Fast Fluid Dynamics (FFD) method as an intermediate approach between nodal models and CFD, providing much richer flow information while being 50 times faster than CFD for conducting faster-than-just-in-time flow simulations for emergency management in buildings. ...
The present state of research in computational fluid dynamics (CFD) is marked by an ongoing process of refining numerical methods and algorithms with the goal of achieving accurate modeling and analysis of fluid flow and heat transfer phenomena. Remarkable progress has been achieved in the domains of turbulence modeling, parallel computing, and mesh generation, resulting in heightened simulation precision when it comes to capturing complex flow behaviors. Nevertheless, CFD faces a significant challenge due to the time and expertise needed for a meticulous simulation setup and intricate numerical techniques. To surmount this challenge, we introduce paint2sim—an innovative mobile application designed to enable on-the-fly 2D fluid simulations using a device’s camera. Seamlessly integrated with OpenLB, a high-performance Lattice Boltzmann-based library, paint2sim offers accurate simulations. The application leverages the capabilities of the Lattice Boltzmann Method (LBM) to model fluid behaviors accurately. Through a symbiotic interaction with the open-source OpenCV library, paint2sim can scan and extract hand-drawn simulation domains, affording the capability for instant simulation and visualization. Notably, paint2sim can also be regarded as a digital twin, facilitating just-in-time representation and analysis of 2D fluid systems. The implications of this technology extend significantly to both fluid dynamics education and industrial applications, effectively lowering barriers and rendering fluid simulations more accessible. Encouragingly, the outcomes of simulations conducted with paint2sim showcase promising qualitative and quantitative results. Overall, paint2sim offers a groundbreaking approach to mobile 2D fluid simulations, providing users with just-in-time visualization and accurate results, while simultaneously serving as a digital twin for fluid systems.
... Solmaz and Gerven [10], Venn et al. [11], and Lin et al. [12] conducted their XR-CFD studies just in visualization aspects, but not in interactive simulation aspect due to its need of separated specialized authoring tools as well as high computational and network costs. Especially in the aspect of computational modeling technique, several loose approximation models proposed by Stam [13], Nguyen et al. [14], Bridson [15] et cetera have been successfully used for interactivity and efficiency. They just employed inviscid and incompressible fluid model. ...
An interactive scientific simulation and visualization pipeline in mixed reality was developed to explore how scientific simulation could proceed in coming workplace. The main technical concern is fast scientific simulator and visualizer in mixed reality with hand gesture. Framework for scientific simulation and visualization was implemented on the server side based on deep learning based fast simulation model and open source visualization toolkit, VTK, respectively. Airfoil aerodynamics was selected as a target scientific simulation task since it is a representative and critical problem in scientific fluid simulation determining the whole aero-product’s performance and shows highly nonlinear physics even by small shape deviation and so high computational cost. Therefore, the fast simulator was firstly developed by carefully developing convolutional neural network-based model, using U-net, which learned corresponding high accuracy simulation dataset. All scenes which are supposed to be seen on the mobile device side were authored in a widely used content development tool called Unity3D with a mixed reality toolkit, MRTK for Microsoft Hololens 2. User interface and user experience were created based on a scenario of a drone rotor which demonstrates running and analyzing the simulation results using designed interaction gestures. Finally, several discussions were derived through the implemented digital twin pipeline. It was concluded, with an adequately implemented mixed reality tools, that scientific researchers or industrial engineers will be helped not only in individual analysis works due to the delicacy of hands and fingers and real time interaction in mixed reality, but also, as a future study, in cooperative analysis works.
... To create physically realistic simulations we rely on the Navier-Stokes equations to help us describe the flow of incompressible fluids. Attempting to model every particle in a smoke plume would achieve the desired physical realism, but is computationally unfeasible, and as such, fluid solvers such as the one proposed by Stam et al. [20] work with density grids. ...
... Creating a simulation of a physically realistic flame that produces smoke is a challenging task, made only more challenging the closer one tries to get to the real world ideal. Attempting to model that physical ideal would take a fluid solver [20] with an infinitely large grid size, which is simply a computational impossibility. Because of this limitation, different methods have been proposed to combat it, such as improving the accuracy of the advection step [11], adding post-processing noise to a coarse base simulation [13] or speeding up the pressure projection step [1]. ...
The following work explores and compares the differences between rendering fire and smoke simulations in high resolution vs rendering these same simulations in low resolution and using deep learning based neural networks to up-scale the output via super-resolution. Several simulations are created at different levels of detail, both with and without post-processing noise added to them. The simulations are then rendered in both high and low resolutions, the lower of which is used for the super-resolution step. The results are then compared in terms of quality and time cost, to determine whether such a computationally expensive task can be improved with deep learning methods. The evaluation shows that using low resolution inputs does not create comparable results to classic high resolution renders, however using a high resolution render of a lower detail simulation creates similar results to high resolution renders of more detailed simulations.
... The ViFlow system employs a particle simulation dancer's can interact with by means of shapes attached to hands and torso with which particles collide [19]. The piece Encoded employs an interactive fluid dynamics simulation to which the dancer's movements are added as velocity-based perturbations [87]. ...
Generative Art is a creative approach that has found applications in several artistic disciplines. In some of these disciplines, formalization has historically played an important role, which predisposes them for employing generative methods. In dance, the relationship to Generative Art is less obvious and the role of formalization is more contested than in other disciplines. This paper tries to contribute to an understanding of the specific role that Generative Art currently plays in dance. It does so by proposing a taxonomy of topics that cover both common and dance specific aspects of Generative Art. This taxonomy is used for comparing a wide diversity of generative works that have been created in the context of dance. CCS CONCEPTS • Applied computing → Performing arts; • Human-centred computing → Interactive systems and tools.
... When the diffusion ratio is greater than 0, some properties will spread out from the cell. Based on the Gauss-Seidel iterative method, we use an implicit method that follows Stam (2003) to compute the diffusion coefficient. This solution can be extended to a fluid in three dimensions, and the results can be obtained through multiple iterations. ...
This paper describes the development of a three-dimensional (3D) physics-based fire simulation model that employs the incompressible Navier–Stokes equations to realistically emulate the combustion process. Then, we animate the 3D interactive burning processes by rendering a flame under a set of influence factors and with various solid boundaries and obstacles. It is insufficient to simply create a virtual reality-based fire model. Instead, evaluating the similarity and accuracy of the models requires data processing for the virtual flames. In this paper, detailed data are extracted from the simulation results to compute the fire’s geometrical features and the distribution of the density and velocity fields. Using methods for video-based fire detection, some visual features of the simulated-fire videos are extracted and compared with those of real fires. The results show the capability of the physics-based fire model in representing some features of real flames. The proposed quantitative analysis of virtual flames serves to evaluate the similarity between a virtual and a real fire.
... The leading concept for the latter ones is smoothed particle hydrodynamics (SPH), which was well summarized by Ihmsen et al. [11]. Well known is also the work of Jos Stam, who eventually presented a convincing real-time fluid solver [21] which combined both approaches. ...
... In order to keep the performance reasonable high, we use an iterative approach to calculate the wind directions instead of applying a computationally expensive fluid dynamics solver. Our method is a simplified version of the semi-Lagrangian scheme [21]. We dropped the diffusion process and the pressure calculations as we handle these separately in a later pipeline step. ...
Advances in computer technology and increasing usage of computer graphics in a broad field of applications lead to rapidly rising demands regarding size and detail of virtual landscapes. Manually creating huge, realistic looking terrains and populating them densely with assets is an expensive and laborious task. In consequence, (semi-)automatic procedural terrain generation is a popular method to reduce the amount of manual work. However, such methods are usually highly specialized for certain terrain types and especially the procedural generation of landscapes composed of different biomes is a scarcely explored topic. We present a novel system, called AutoBiomes, which is capable of efficiently creating vast terrains with plausible biome distributions and therefore different spatial characteristics. The main idea is to combine several synthetic procedural terrain generation techniques with digital elevation models (DEMs) and a simplified climate simulation. Moreover, we include an easy-to-use asset placement component which creates complex multi-object distributions. Our system relies on a pipeline approach with a major focus on usability. Our results show that our system allows the fast creation of realistic looking terrains.
... Ces équations entrent également dans l'étude de la circulation du sang dans nos artères, dans la simulation des trajectoires d'air autour d'une aile d'un avion et dans la simulation des tourbillons. Ces équations sont même utilisées dans les jeux vidéos pour améliorer le réalisme de certaines scènes [76]. ...
In this manuscript, we study the three-dimensional stationary Stokes equations set in a exterior domain. The problem describes the flow of a viscous and incompressible fluid past a bounded obstacle. The distinctif feature here relies on the fact that the obstacle is assumed to a rough boundary. As a result, the fluid may slip on the boundary of the obstacle and, to take into account this property, we use the Navier boundary conditions. On the one hand, They model the impermeability of the obstacle, and on the other hand, the fact that the tangential component of the fluid velocity on the obstacle is proportional to the stress tensor. This problem has been well studied when set in a bounded domain. The standard Sobolev spaces provides, in this case, an adequate functional framework for a complete study. Since in our case, the domain is unbounded, these spaces are not adapted since it is necessary to describe the behaviour of the solutions to infinity. Therefore, we choose to set the problem in weighted Sobolev spaces where the weights describe the behaviour at infinity of the function (growth or decay).In this work, we first start by performing the mathematical analysis in the Hilbert setting. The key point here is to establish variant weighted Korn’s inequalities in order to get the coercivity of the bilinear form associated to the variational formulation. Next, we proved the existence, uniqueness of strong and very weak solutions. Finally, we study the extension of some of thses results to a weightedL^p-theory.
... In this paper, we utilise a reasonably accurate simulation of the phenomenon [29] but exploit simplifications inherent to the problem, such as that the dispersion process can be modelled on the 2-d plane 1 along which the point measurements are also being taken. Moreover, the process of dispersion is shift invariant [30], so that a single large simulation can be performed online, from which the flow patterns for different locations can be easily computed. ...
... Fluid simulator Another requirement for our method is having access to a simulator oracle g. There are many ways to express a physical model of fluids, but we opt to use a stable Navier-Stokes solver due to its efficiency and ease of implementation [29]. The solver is realized by dividing the space into voxels and iteratively updating the velocity and density. ...
Sensors are routinely mounted on robots to acquire various forms of measurements in spatio-temporal fields. Locating features within these fields and reconstruction (mapping) of the dense fields can be challenging in resource-constrained situations, such as when trying to locate the source of a gas leak from a small number of measurements. In such cases, a model of the underlying complex dynamics can be exploited to discover informative paths within the field. We use a fluid simulator as a model, to guide inference for the location of a gas leak. We perform localization via minimization of the discrepancy between observed measurements and gas concentrations predicted by the simulator. Our method is able to account for dynamically varying parameters of wind flow (e.g., direction and strength), and its effects on the observed distribution of gas. We develop algorithms for off-line inference as well as for on-line path discovery via active sensing. We demonstrate the efficiency, accuracy and versatility of our algorithm using experiments with a physical robot conducted in outdoor environments. We deploy an unmanned air vehicle (UAV) mounted with a CO2 sensor to automatically seek out a gas cylinder emitting CO2 via a nozzle. We evaluate the accuracy of our algorithm by measuring the error in the inferred location of the nozzle, based on which we show that our proposed approach is competitive with respect to state of the art baselines.
... Third, the system adds stripes to the fins and tail by Region-based Line Field Design Using Harmonic Functions [Yao et al. 2012], details can be seen on Figure 4. Forth, for the contours of the koi, our system applies the Suggestive Contours algorithm [DeCarlo et al. 2003] [Proenca et al. 2008] to find both outlines and inner contours. Fifth, we apply our preprocessed stroke pattern, generated by the Navier-Stokes equation, [Stam 2003] on the outlines, inner contours, and stripes of the koi. The result can be seen in Figure 2(d). ...
... For the user-interactive ripples, the ripples are generated by Mass-Spring Model Ripple Simulation [Zhang and Yang 2010] upon userinputs, as shown in Figure 5(a). In order to achieve the Chinese painting look, the Navier-Stokes equation [Stam 2003] is applied to the ripples. The final appearance of the ripples can be seen on Figure 5(b). ...
Dynamic Chinese Painting is a rendering style for animations. However, the production method today which relies on the artists to generate suitable textures and sceneries is time-consuming. In this paper, we propose a system to generate a realistic appearance of Yangzhou School Chinese Painting koi (decorative carp) animation in 3D space. Our system includes repainting the original texture, enhancing the contour of the 3D input models, and adding fin-stripes to fit the target style. For the interactive part, our system not only allows users to add ripples to the scene as the user-scene interaction but also generates water streamline as the koi-scene interaction. Through our system, a new Dynamic Chinese Painting scene can be automatically generated based on the 3D input models and user-inputs.
