Emmanuelle Darles

A Survey of Ocean Simulation and Rendering Techniques in Computer Graphics

This paper presents a survey of ocean simulation and rendering methods in computer graphics. To model and animate the ocean’s surface, these methods mainly rely on two main approaches: on the one hand, those which approximate ocean dynamics with parametric, spectral or hybrid models and use empirical laws from oceanographic research. We will see that this type of methods essentially allows the simulation of ocean scenes in the deep water domain, without breaking waves. On the other hand, physically‐based methods use Navier–Stokes equations to represent breaking waves and more generally ocean surface near the shore. We also describe ocean rendering methods in computer graphics, with a special interest in the simulation of phenomena such as foam and spray, and light’s interaction with the ocean surface.

Hybrid Physical -- Topological Modeling of Physical Shapes Transformations

This work is grounded on the idea that physical and topological properties are complementary features in visual objects and their transformations. Indeed, physical properties delimit the quality of the dynamics whereas topological properties focus on the spatial organization. In this paper, we propose a new cooperation between mass-interaction models for modeling motion and topological maps for modeling topology. We experiment it through two basic examples: the tearing of a fabric and the breaking-sticking of a cohesive material.

Topology-based Physical Simulation

This paper presents a framework to design mechanical models relying on a topological basis. Whereas naive topological models such as adjacency graphs provide low topological control, the use of efficient topological models such as generalized maps guarantees the quasi-manifold property of the manipulated object: Topological inquiries or changes can be handled robustly and allow the model designer to focus on mechanical aspects. Even if the topology structure is more detailed and consumes more memory, we show that an efficient implementation does not impact computation time and still enables real-time simulation and interaction. We analyze how a simple mass/spring model can be embedded within this framework.

A New Force Model for Controllable Breaking Waves

(a) Multiple waves displayed from the side (b) from above (c) Waves interacting with static blocks Figure 1: Breaking waves obtained with our model. An ocean scene with multiple waves is displayed (a) from the side and (b) from above; (c) and breaking waves and backflow waves colliding with blocks. The waves combine naturally, while each of them is modeled with its specific parameters (height, width, speed, orientation, crest slope, breaking time). Abstract This paper presents a new method for controlling swells and breaking waves using fluid solvers. With conventional approaches that generate waves by pushing particles with oscillating planes, the resulting waves cannot be controlled easily, and breaking waves are even more difficult to obtain in practice. Instead, we propose to use a new wave model that physically describes the behavior of wave forces. We show that mapping those forces to particles produces various types of waves that can be controlled by the user with only a few parameters. Our method is based on a 2D representation that describes wave speed, width, and height. It handles many swell and wave configurations, with various breaking situations. Crespin / A New Force Model for Controllable Breaking Waves

Accelerated Simulation of Brittle Objects for Interactive Applications

This article presents a model that aims at computing deformation and simulating fractures. To allow the use of linear elasticity in small displacements for deformation of a brittle object while still enabling any rigid motion, a rigid reference is computed using the Shape Matching method and all displacements are evaluated with respect to this reference. Fractures are handled using a stress tensor computed at each vertex of the object’s 3D mesh. Some accelerations are proposed, that allow a faster determination of fracture areas and a fast processing of new connected components.

A Topological-Geometrical Pipeline for 3D Cracking-like Phenomena

Animation of one-to-many phenomena (fractures, tears, breaks, cracks...) is challenging. This article builds over recent works that proposed a 3-stages modelling and simulation pipeline, made of a cascade of models: geometry-free physical model → explicit modelling of the evolving topology → geometrical model. On the Physics’ side, in the framework of masses-interactions network modelling, the article extends the recent Splitting-MAT method, where the physical splits occur onto the material points, toward 3 dimensional volume models. Downstream, it introduces a topo-geometrical pipeline adapted to this upstream split-on-the-masses property. Experiments, and analysis of the complexity of the topo-geometrical part, show that, while offering constructible and manageable means, separating Physical, Topological and Geometrical aspects in the 3stages pipeline enables a rich variety of one-to-many dynamics, with good efficiency.

A New Modeling Pipeline for Physics-Based Topological Discontinuities: The DYNAMe Process

This work introduces a new modeling/simulation pipeline for handling topological discontinuities as they appear in fracturing, tearing or cracking phenomena. This pipeline combines, in a cascade (1) physics-based particle modeling, which enables a wide variety of temporal phenomena, including physics state changes; (2) explicit topological modeling, which makes it possible to manage a wide variety of shape-independent topologies and robust topological transformations; and, in between, free, non predetermined association, enabling topological transformations all along the animation under control of the point-based movement produced by the physics-based model.

Mapping Volumetric Meshes to Point-based Motion Models

Particle-based models produce various, flexible and optimized animations. Intrinsically, they are neither based on a boundary representation nor on volumetric objet meshed representations. Thus, they raise rendering issues since they do not contain enough geometrical information and do not even provide an underlying spatial topology. Consequently, various geometrical shapes can be used to render a motion produced by a meshless model, leading to different visual interpretations. To our knowledge, there is no generic methods to associate any set of points in motion with a topology-based geometric model. In this paper, we propose a framework to map arbitrary volumetric meshes to arbitrary point-based motions and to control the topological changes. Therefore, from only one motion description, different visual results can be obtained. This framework breaks down into three distinct processes: a particles to vertices mapping, the definition of a motion function and the definition of topological modifications and events triggering them. We show how the manipulation of these parameters allows to experiment different mappings for a particular motion and that our framework includes most of previous known mappings.

A particle-based method for large-scale breaking waves simulation

We address in this paper the problem of particle-based simulation of breaking waves. We present a new set of equations based on oceanographic research which allow us to deal with several types of breaking waves and multiple wave trains with full control over governing parameters. In order to reduce computations in non-significant areas, we also describe a simple and efficient multiresolution scheme, controlled using the properties of our breaking wave model.