Abstract: A silhouette edge is rendered in a display format that corresponds to a geometry of a three-dimensional model. An angle between the silhouette edge and an adjacent silhouette edge in the image is determined, and the display format for the silhouette edge is selected based on the angle. Texture mapping is used to generate the silhouette edge in the display format.

Abstract: Different limit surfaces are derived from the same initial arbitrary polygon mesh by sequentially combining different subdivision rules. This added freedom allows for the more efficiently modeling of objects in computer graphics including objects and characters with semi-sharp features.

Abstract: A mobile data apparatus equipped with a display is improved in both operability and mobility. A frame difference detector 4 is provided for detecting a change in image between two consecutive frames transferred from a pair of frame buffers 2 and 3. The change in image is used in a barycenter calculator 5 for calculating the barycenter of each image. The barycenter is saved in a barycenter coordinates memory 6. A camera movement calculator 6 calculates a difference between the barycenter saved in the barycenter coordinates memory 6 and a reference barycenter saved in a barycenter coordinates memory 7 to determine a horizontal pixel shift DPx and a vertical pixel shift DPy of a camera. In response to those shifts, a horizontal angle-of-rotation controller 9 and a vertical angle-of-rotation controller 10 determine angles of rotation along the horizontal and vertical directions hx and hy for a virtual camera 11. Accordingly, the view of the virtual camera 11 can duly be controlled.

Abstract: An image processing system in which the vertices of an object contained within an image are analyzed using a singular value decomposition (SVD) method is disclosed. The use of the SVD allows the original vertices data to be reduced through filtering or truncating the singular values associated with the image. In addition, the reduced vertices data and the associated SVD matrices allows for efficiently streaming video data.

Abstract: A mobile data apparatus equipped with a display is improved in both operability and mobility. A frame difference detector 4 is provided for detecting a change in image between two consecutive frames transferred from a pair of frame buffers 2 and 3. The change in image is used in a barycenter calculator 5 for calculating the barycenter of each image. The barycenter is saved in a barycenter coordinates memory 6. A camera movement calculator 6 calculates a difference between the barycenter saved in the barycenter coordinates memory 6 and a reference barycenter saved in a barycenter coordinates memory 7 to determine a horizontal pixel shift DPx and a vertical pixel shift DPy of a camera. In response to those shifts, a horizontal angle-of-rotation controller 9 and a vertical angle-of-rotation controller 10 determine angles of rotation along the horizontal and vertical directions hx and hy for a virtual camera 11. Accordingly, the view of the virtual camera 11 can duly be controlled.

Abstract: A limitation of modeling three-dimensional objects in a virtual world occurs when a user moves within the world and still wishes a certain target object to remain visible, instead of being occluded by other surfaces in front of the object along the direction the user is viewing the target object. Therefore, in accordance with the principles of the invention, a target object which is to remain visible, regardless of the user's view point, is duplicated and modeled as being located a close distance to the user, but scaled accordingly to make the cloned object appear to be farther away from the viewpoint of the user and co-located with and of the same size and orientation as the target object. In this manner, as the user moves within the world, surfaces located in front of the target object do not occlude the cloned object from the user's view, thus appearing to give that same property to the target object.

Abstract: A storage unit includes a part table which stores animation parts indicative of actions and expressions of a person divided into a plurality of frames and which further stores attribute values of the animation parts. An entry unit is used to enter the attribute values of the animation parts in accordance with the proceeding steps of a story. An arithmetic unit selects animation parts from the storage unit using the attribute values entered through the entry unit and creates an animation in conformity with the story.

Abstract: A graphics pipeline system is provided for graphics processing. Such system includes a transform module adapted for receiving vertex data. The transform module serves to transform the vertex data from a first space to a second space. Coupled to the transform module is a lighting module which is positioned on the single semiconductor platform for performing lighting operations on the vertex data received from the transform module. Also included is a rasterizer coupled to the lighting module and positioned on the single semiconductor platform for rendering the vertex data received from the lighting module. During use, an antialiasing feature is implemented to improve a quality of the graphics rendering.

Abstract: In order to visually display a spatial distribution of a multi-value function represented by nine components of a two-order, three-dimensional tensor, a distribution function of an anisotropic fine face distribution is set to have a form of a second-order curved surface of an anisotropic ellipsoid associated with discrete sampling points, the intensity of the scattering light is calculated from a contribution of a bisector-direction vector between a light-source-directed vector and a line of sight vector (in a reflection light intensity calculating part to simulate the light scattering. With respect to each calculation point having a tensor, the intensity of the scattering light is integrated according to a rendering light beam equation with a light transmittance weighted (in a light beam integration part).

Abstract: A software visualization tool enables a user to observe the inner workings of a computer program (e.g., a Java program) while it executes. An original program is augmented with calls to a shadow machine at locations where the original program modifies a data structure maintained by the data program or updates values in a value set maintained within the context of the data structure. A duplicate copy of the data structure and the value set for the augmented program is maintained during execution of the computer program by a shadow machine. A display engine accesses the duplicate copy of the data structure and the value set, and displays a visual representation of the duplicate copy on a display device so as to facilitate visualization of the execution of the augmented program in a dynamic and near real-time manner.

Abstract: In an apparatus and method for representing computer-modeled objects comprising a plurality of adjacent polygon surfaces, the image impression in the corner points of the polygon surfaces is calculated according to a local illumination model, whereas the image impression of the remaining pixels is interpolated according to the Gouraud shading method, and a color or brightness value is additionally calculated for simulating a surface structure for each pixel, the value taking into consideration the influence of the surface structure on the image impression without necessitating a complicated calculation of the local illumination model in each pixel.

Abstract: Real-time rendering of large-scale surfaces with locally adapting surface geometric complexity according to changing view parameters. A view-dependent progressive mesh (VDPM) framework represents an arbitrary triangle mesh as a hierarchy of geometrically optimized refinement transformations, from which accurate approximating meshes can be efficiently retrieved. The VDPM framework provides temporal coherence through the run-time creation of geomorphs. These geomorphs eliminate “popping” artifacts by smoothly interpolating geometry. The geomorphs utilizes output-sensitive data structures to reduce memory requirements.

Abstract: A key frame animation processing method includes steps of outputting information indicating a time for presenting a key frame for every key frame, and outputting state information of each object in correspondence to the time information of the key frame for every key frame, thereby reducing the quantity of outputted data. The data may be outputted to either a storage medium or a communication medium. This method may include a step of outputting link information as to each object and another object in another key frame in correspondence to the state information of the object. The state information may include only the quantity of a state change. Also disclosed is a method of displaying a key frame animation in real time.

Abstract: An interactive virtual endoscopy system 10 includes a CT scanner 12 or other non-invasive examination apparatus which examines an interior region of a subject 14 in an examination region 16 and generates data indicative thereof. The data is stored in a volume image data memory 20. Using a sequence generating computer 22, a human operator 24 generates a sequence of sphere-mappable panoramic views of selected portions of the CT data along a viewpath in the patient 14. The sequence generating computer includes a view renderer 182 for rendering a plurality of views which in total cover the entire visual space about a viewpoint on the viewpath within the subject. A view compositer 240 combines the plurality of views into a full image covering the entire visual space about the viewpoint. The sequence is transferred to a server 26 which processes the data and makes it available for remote access.

Abstract: A computer-based system for designing and using three-dimensional environments over a bandwidth limited network such as the Internet. The system allows an environment to be specified as a series of two-dimensional grids of text characters. Each character occupies a single grid position and represents an object in the environment. Objects can be given characteristics such as texture maps, and associated images and sounds that are triggered by events such as a user approaching the object. An object or image can be a hyperlink so that, when clicked or moved upon, the user is transported to a new location. A basic set of objects and media (images and sounds) is provided so that a designer of an environment does not have to perform low-level three-dimensional modeling. Objects can behave differently when placed near one another. For example, walls fuse together to provide a longer wall.

Abstract: To create images where the texture of a model changes with a relatively small image data volume three-dimensional model wherein texture changes from one texture to another is obtained by repeatedly superimposing images of the two three-dimensional models while changing the transparency of respective images complementarily.

Abstract: A method and apparatus for creating a blend from one arbitrary edge (20) to a second arbitrary edge (21) in a computer graphic image creation is disclosed. A color along each of the edges (20,21) is determined. A parametric equation is then formed for a color of each pixel (23) within the area bounded by the edges (20,21), and the parametric equation is solved to derive a color for each of the pixels (23). In another aspect, after determining the color along each edge (20,21), each of the edges (20,21) is vectorised into corresponding line segments (31 to 34). Pairs of the line segments (31 to 34) are then matched to form polygons (29) having a defined color at each of vertices, A color is then determined for each pixel of the polygon (29) from the defined colors of the vertices.

Abstract: A limitation of a three-dimensional world in which objects in the distance may be represented in only two dimensions as a video on a screen occurs when a computer graphic object represented by computer graphics, e.g., in front of, to the side, above, or below the video screen, undergoes a trajectory that takes it to a location in the world that is not represented as computer graphics, but instead is within the field represented by the video, because such an object would disappear from view by the viewer. This limitation is overcome by having such an object be represented as video on the screen, rather than computer graphics. Thus, the computer graphics object “goes into the video” as video and remains visible to a viewer in front of the video screen, rather than becoming invisible because it is blocked from view by the video screen if it were to be generated at its proper location using computer graphic techniques.

Abstract: Users control the movement of avatars through a virtual space. The system keeps track of the location of the avatars and forms a picture of the virtual space for each user. Users can select themselves to act as a cameraman. The picture of the virtual space around the location of the avatar is then transmitted separately to passive viewers via a broadcast channel. Conversation between the cameraman and other users is broadcast also via the broadcast channel. The picture broadcast has a visual detail which is finer than that of the pictures received by the various users individually; movements are reproduced therein in interpolated form and are adapted in such a manner that the avatars in the broadcast signal arrive at landmarks, in time.

Abstract: In a method for determining a data value for a target pixel in a destination image based on data values for pixels in a source image, with the destination image being scaled relative to the source image, calculating a position in the source image based on position of a target pixel in the destination image, testing the presence of a diagonal gradient in the source image at the position determined in the calculating step, the testing step testing for the presence of a diagonal gradient by reference to values of pixels in the source image that surround the position calculated in the calculating step, responsive to the presence of a diagonal gradient in the testing step, calculating a data value for the target pixel based on interpolation of data values for diagonally-adjacent pixels in the source image, and responsive to the absence of a diagonal gradient in the testing step, calculating a data value for the target pixel based on interpolation of data values for at least all four surrounding pixels in the sourc