International Business Machines Corporation

James Thomas Klosowski

Frank Suits

William Pettit Horn

Claudio T Silva

Gerard Lecina

Shawn Anthony Hall

Paul Borrel

William Louis Luken

Ioana Maria Martin

David Patrick Luebke

Geetika Tewari Lakshmanan

Jai P Menon

Gabriel Taubin

Christopher J Morris

Fausto Berdardini

Jihad El-Sana

Julia Anatolyevna Valuyeva

Peter Lawrence Williams

Joao Luiz Dihl Comba

Joseph Shannon Baird Mitchell

A method and system for managing graphics load balancing strategies are disclosed. The method comprises using a plurality of rendering servers to render a multitude of graphics frames for a display device, wherein each of the rendering servers has an associated workload; identifying a plurality of load balancing strategies for balancing the workloads on the rendering servers; selecting one of the load balancing strategies; and using the selected one of the load balancing strategies to balance the workloads on the rendering servers. One or more defined metrics are monitored; and in response to a defined changed in said one or more defined metrics, another one of the load balancing strategies is selected and used to balance the workloads on the rendering servers. In one embodiment, the load balancing policy can be changed in real-time during the course of an application session.

Managing graphics load balancing strategies

A system, method and computer program product for parameterizing a 3D surface of genus g and arbitrary topology to the 2D plane. The method includes receiving an input 3D mesh of genus g; computing n discrete harmonic one-forms on the surface of the mesh to obtain n numbers for every edge in the graph represented by an input mesh; selecting two compatible linearly independent discrete harmonic one-forms for parameterization; assigning an arbitrary vertex in the mesh as an origin, and then summing each of the two harmonic one-forms along the edges of the graph using a traversal strategy to form 2D coordinates for every other vertex of the mesh to form the 2D parameterization of the 3D surface; detecting all singularities in the 2D parameterization; and, removing faces connected to the singularity and parameterizing the vertices of the removed faces using a 2D parameterization method, and connecting these vertices using a 2D polygonalization algorithm. Using the 3D mapping methodology employed, a correspondence function is computed that can be used to match two 3D object surfaces and repair meshes of arbitrary genus.

System and method for establishing correspondence, matching and repairing three dimensional surfaces of arbitrary genus and arbitrary topology in two dimensions using global parameterization

Vertices of an annotation are projected onto a surface of a (2D or 3D) model and reconnected to preserve the original appearance of the annotation. The result of our method is a new set of geometry for the annotation that geometrically conforms to the surface of the model. A plane is defined by the midpoint of the original line segment and the two projected vertices. This plane is used to create the new line segments that conform to the surface, e.g. by doing a 'surface walk' between the projected points along the line defined by the intersection of the plane and the surface.

Apparatus, system, and method for draping annotations on to a geometric surface

A technique and system for selecting level-of-detail representations of geometric models to be rendered within an image processing system. For each geometric model, fractional visibility estimations are computed, thereby ranking how likely it is that a model is visible. Using these rankings, an appropriate level-of-detail for each geometric model is selected to optimize the rendering process by reducing the number of primitives that need to be rendered, while preserving the quality of the final image produced and displayed upon the screen. Visibility estimates for the geometric models are summed to produce a number which is then used to scale the number of primitives used in the level-of-detail representation of the geometric models.

System and process for level of detail selection based on approximate visibility estimation

An occlusion culling method for image processing and system therefor. The culling method first determines which polygons are hidden, or occluded, by other objects. These hidden or occluded polygons will not contribute to the final scene and, therefore, need not be rendered. In the first major step, the input models are preprocessed to build a hierarchical data structure which is as an approximation to the input models. Simple polygonal occluders are determined for substitution in place of the complex image geometry in successive visibility queries. Once the first preprocessing step is complete, the second step may be executed at run-time, while a user is inspecting, or visualizing, the input. In the second step, the occluders, determined in the first step, are used to selectively forego rendering shapes or shape portions that are unseen from the current viewpoint.

Accelerated occlusion culling using directional discretized occluders and system therefore

A system simplifies a geometric model to accelerate the rendering of the geometric model. A surface description of the geometric model is stored in one or more of the system memories. A wire-frame description of the geometric model is also stored. A surface simplification process alters the surface description to create an approximation of the original surface description. A wire-frame draping process drapes the wire-frame description onto the simplified surface and simplifies one or more of the line segments that are draped, the simplified line segments and the simplified surface description are rendered onto a display device.

System and method for the coordinated simplification of surface and wire-frame descriptions of a geometric model

A system for providing interactive views of 3-dimensional models with surface properties is disclosed. The system provides a compact representation of a 3D model and its surface features and provides for efficiently viewing and interacting with the model using dynamically switched texture maps. The compact representation is beneficial for transmission of the model across a network as well as for local storage of the model in the computer memory. The dynamically switched texture maps allow for more accurate surface details on the 3D model, as well as speedy interaction between a user and the 3D model.

Dynamic view-dependent texture mapping

A computer system and method for simplifying annotations on a surface includes a tolerance process creates a tolerance window lying in a plane and centered on the query point at the end of a test edge and perpendicular to the test edge. The tolerance window is defined by a first tolerance that specifies an amount of first error measured by the distance from a simplified path to an original path in the plane tangent to the surface, and a second tolerance that specifies an amount of second error measure by the distance from the simplified path to the original path in the plane perpendicular to the surface. The original path is a set of original edges on the surface. A point projection process that projects a second end point of one or more second edges onto the plane of the tolerance window. The second edges has the query point also as an end point where the projection of the second end points are projected points. An invalidation process marks an edge as &ldquo;invalid&rdquo; and removes it both from the list of simplifiable edges and the final list of simplified edges. A selection process selects one of the second end points that has a projected point within the plane containing the tolerance window, deletes the query point, and connects the origin point to the respective second end point, to create a new edge that replaces the test edge and respective second edges in the set of original edges to create the simplified path.

Apparatus, system, and method for simplifying annotations on a geometric surface

A Prioritized-Layered Projection (PLP) method and system for optimizing the rendering high-depth complexity scenes. Objects in each frame are prioritized with an estimation of the visible set within the rendered frame. A priority order for only the primitives in visible sets within the frame are computed &ldquo;on demand&rdquo; to maximize the likelihood of rendering visible primitives before rendering occluded ones. For a fixed budget, such as, time or number of triangles, rendering is constrained to a geometry budgeted priority. The method includes two main steps: an occupancy-based tessellation of space; and a solidity-based traversal algorithm. By first computing an occupancy-based tessellation of space, more cells result where there are more geometric primitives. In this spatial tessellation, each cell is assigned a &ldquo;solidity&rdquo; value directly proportional to its likelihood of occluding other cells. In its simplest form, a cell's solidity value is directly proportional to the number of primitives contained within the cell. Cells are marked for projection, and the geometric primitives contained within the marked cells are rendered. Cell solidity, and other view-dependent information determines the order cells are rendered.

System, program product and method of rendering a three dimensional image on a display

A system and method for seamlessly combining client-only rendering techniques with server-only rendering techniques. The approach uses a composite stream containing three distinct streams. One stream is available to send geometry from the server to the client. Another stream contains video with transparent pixels that allow the client-rendered object to appear in the context of the server rendered objects. The third stream contains camera information. The invention can satisfy a number of viewing applications. For example, initially the most relevant geometry can stream to the client for high quality local rendering while the server delivers renderings of less relevant geometry at lower resolutions. After the most relevant geometry has been delivered to the client, the less important geometry can be optionally streamed to the client to increase the fidelity of the entire scene. In the limit, all of the geometry is transferred to the client and the situation corresponds to client-only rendering system where local graphics hardware is used to improve fidelity and reduce bandwidth. Alternatively, if a client does not have local three-dimensional graphics capability then the server can transmit only the video of the server-rendered object and drop the other two streams. In either case, the approach also permits for a progressive improvement in the server-rendered image whenever the scene becomes static. Bandwidth that was previously used to represent changing images is allocated to improving the fidelity of the server-rendered image whenever the scene becomes static.

Method and apparatus for delivering 3D graphics in a networked environment using transparent video

A system and method for seamlessly combining client-only rendering techniques with server-only rendering techniques. The approach uses a composite stream containing three distinct streams. Two of the streams are synchronized and transmit camera definition, video of server-rendered objects, and a time dependent depth map for the server-rendered object. The third stream is available to send geometry from the server to the client, for local rendering if appropriate. The invention can satisfy a number of viewing applications. For example, initially the most relevant geometry can stream to the client for high quality local rendering while the server delivers renderings of less relevant geometry at lower resolutions. After the most relevant geometry has been delivered to the client, the less important geometry can be optionally streamed to the client to increase the fidelity of the entire scene. In the limit, all of the geometry is transferred to the client and the situation corresponds to client-only rendering system where local graphics hardware is used to improve fidelity and reduce bandwidth. Alternatively, if a client does not have local three-dimensional graphics capability then the server can transmit only the video of the server-rendered object and drop the other two streams. In either case, the approach also permits for a progressive improvement in the server-rendered image whenever the scene becomes static. Bandwidth that was previously used to represent changing images is allocated to improving the fidelity of the server-rendered image whenever the scene becomes static.

Methods and apparatus for delivering 3D graphics in a networked environment

The visibility ordering of polyhedral cells is efficiently determined by building an ordering graph, comprising oriented edges between two cells. Each edge (A,B) corresponds to the fact that cell A has to be projected, or rendered, before B. A set of ordering relations and rules that can be shown to generate, if one exists, a global ordering of the polyhedral cell complex. Three different types of edges are used to accomplish this: MPVO, BSP and PPC edges. MPVO edges exist between two cells that share a face. To define the BSP edges, a BSP-tree of the boundary faces of the cell complex is constructed. During this construction, some of the boundary faces of the cells will be 'cut' by the BSP-tree 'extended' faces, into multiple pieces. If C is the boundary cell, and c&prime;, c&Prime;, and so on, are the pieces of its boundary faces, the BSP_edge (c&prime;, C) is defined to mean that cell C can only be projected after c&prime; has been projected by the BSP. A cell C is in the PPC if one of the pieces that compose it, say c&prime;, has been projected by the BSP, but there exist other pieces of cell C that have not been projected. In this case, additional checks are done to ensure cells are not being projected out of order. This is accomplished by performing &lsquo;ray shooting&rsquo;.

Growing community of inventors

Rye, NY, United States of America

James Thomas Klosowski