Jérémie Allard

Image-based collision detection and response between arbitrary volume objects

We present a new image-based method to process contacts between objects bounded by triangular surfaces. Unlike previous methods, it relies on image-based volume minimization, which eliminates complex geometrical computations and robustly handles deep intersections. The surfaces are rasterized in three orthogonal directions, and intersections are detected based on pixel depth and normal orientation. Per-pixel contact forces are computed and accumulated at the vertices. We show how to compute pressure forces which serve to minimize the intersection volume, as well as friction forces. No geometrical precomputation is required, which makes the method efficient for both deformable and rigid objects. We demonstrate it on rigid, skinned, and particle-based physical models with detailed surfaces in contacts at interactive frame rates.

Grimage: markerless 3D interactions

Grimage glues multi-camera 3D modeling, physical simulation and parallel execution for a new immersive experience. Put your hands or any object into the interaction space. It is instantaneously modeled in 3D and injected into a virtual world populated with solid and soft objects. Push them, catch them and squeeze them.

Multicamera Real-Time 3D Modeling for Telepresence and Remote Collaboration

We present a multicamera real-time 3D modeling system that aims at enabling new immersive and interactive environments. This system, called Grimage, allows to retrieve in real-time a 3D mesh of the observed scene as well as the associated textures. This information enables a strong visual presence of the user into virtual worlds. The 3D shape information is also used to compute collisions and reaction forces with virtual objects, enforcing the mechanical presence of the user in the virtual world. The innovation is a fully integrated system with both immersive and interactive capabilities. It embeds a parallel version of the EPVH modeling algorithm inside a distributed vision pipeline. It also adopts the hierarchical component approach of the FlowVR middleware to enforce software modularity and enable distributed executions. Results show high refresh rates and low latencies obtained by taking advantage of the I/O and computing resources of PC clusters. The applications we have developed demonstrate the quality of the visual and mechanical presence with a single platform and with a dual platform that allows telecollaboration.

Net Juggler: running VR Juggler with multiple displays on a commodity component cluster

Net Juggler is an open source library that turns a commodity component cluster running the VR Juggler platform on each node into a single VR Juggler image cluster. Application parallelization is transparent to the user and leads to high performance executions even with limited bandwidth networks.

Implicit FEM Solver on GPU for Interactive Deformation Simulation

Publisher Summary This chapter presents a set of methods to implement an implicit Finite Element solver on the graphics processing units (GPU). The Finite Element Method (FEM) is broadly used to simulate deformable materials in physics simulations. Existing GPU-based methods implement FEM only with explicit time integrators, which are simple and easy to parallelize. However, these methods suffer from stability issues and require very small time steps to simulate stiff materials. The finite element method provides a means for discretizing and solving volumetric models of deformable materials. To parallelize a given set of computations on the GPU it is necessary to extract a massive level of parallelism, on the order of tens of thousands of threads. In many cases, the computations are independent except that they need to scatter their results onto a set of shared variables. This happens for instance when computing FEM elements accumulating forces to, or when computing a sparse matrix–vector product. Such operations can be represented as a graph, where nodes represent shared variables and edges the computations between them. A very common technique to parallelize such a graph is to partition it into a set of subgraphs, each computed by a different processor. To achieve maximum performance it is critical to design data layouts to optimize memory access. A common approach to improve cache efficiency of memory access patterns is to convert large arrays of structures into a structure of arrays. The fastest method to compute the local frame of an element, although not the most accurate one, is to set the origin at vertex p0 and use its adjacent edges.

Asynchronous Preconditioners for Efficient Solving of Non-linear Deformations

In this paper, we present a set of methods to improve numerical solvers, as used in real-time non-linear deformable models based on implicit integration schemes. The proposed approach is particularly beneficial to simulate nonhomogeneous objects or ill-conditioned problem at high frequency. The first contribution is to desynchronize the computation of a preconditioner from the simulation loop.We also exploit todays heterogeneous parallel architectures: the graphic processor performs the mechanical computations whereas the CPU produces efficient preconditioners for the simulation. Moreover, we propose to take advantage of a warping method to limit the divergence of the preconditioner over time. Finally, we validate our work with several preconditioners on different deformable models. In typical scenarios, our method improves significantly the performances of the perconditioned version of the conjugate gradient.

A (Near) Real-Time Simulation Method of Aneurysm Coil Embolization

An aneurysm is an abnormal widening of a blood vessel. As the vessel widens, it also gets thinner and weaker, with an increasing risk of rupture. Aneurysms are essentially found in the aorta, the popliteal artery, mesenteric artery, and cerebral arteries. Intracranial aneurysms are smaller than other types of aneurysm and mostly saccular. Though most patients do not experience rupture, it can lead to a stroke, brain damage and potential death. The mortality rate after rupture is considerably high: the incidence of sudden death was estimated to be 12.4% and death rate ranged from 32% to 67% after the hemorrhage [16]. Each year, over 12,000 people die in the United States due to rupture of intracranial aneurysms [17].

GPU-Based Interactive Simulation of Liver Resection

This GPU-based interactive simulation of laparoscopic liver resection is implemented using the open-source SOFA framework. While similar medical simulators have been developed in the past, this demo relies on advanced methods and the computational power of current GPUs to simulate multiple organs with high-resolution deformations and collisions in real time. It is based on recently proposed methods: high-resolution Finite Element Model (FEM) with implicit time-integration implemented on GPU, volume-contact constraints, an efficient numerical solver based on asynchronous preconditioning, and improvements in visual and haptic rendering. And it uses detailed meshes generated from segmented CT scans to facilitate reproduction of patient-specific scenarios, which is necessary for pre-operative rehearsal of complex or risky medical procedures. These methods allow real-time simulation of all organs in the abdominal cavity using an improved level of precision compared to previous systems. The FEM formulation enables reproduction of specific material properties. Contacts are handled by precise constraints with frictions on detailed surface meshes. Both methods efficiently support topological changes, as demonstrated by performing a resection of a portion of a liver, an important step in surgical procedures performed to remove cancerous tumors. Users can examine the mechanical and collision models, and the generated contacts while the simulated patient is breathing, and manipulate a laparoscopic instrument to navigate through the abdominal cavity, push on organs, and perform a thermal ablation. This project is a result of a collaboration between INRIA and IRCAD within the EU-funded PASSPORT project.

Coupling parallel simulation and multi-display visualization on a PC cluster

Recent developments make it possible for PC clusters to drive multi-display visualization environments. This paper shows how we coupled parallel codes with distributed 3D graphics rendering, to enable interactions with complex simulations in multi-display environments.