Feiner, S.K., Foley, J. D., Hughes, J.F., van Dan, A. Computer Graphics: Principles and Practice. 2nd Edition, Addison Wesley, 1992.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....5.1 Hardware Based Renderer In our implementation of the hardware based renderer, we follow the algorithm outlined in [17] In a first pass, we compute the amount of light that directly reaches each voxel of the grid. This is achieved using a fast Bresenham line drawing voxel traversal algorithm [8]. Initially the transparencies of each ray are set to one (#### ##) Then, each time a voxel is hit, the transparency is computed from the voxel s density: #### ############, where # is the grid spacing. Then the voxel s radiance is set to ###### # ##### ## # #### # #### # while the ....

....interpolations along each coordinate axis. Therefore, it is sufficient to describe the one dimensional case only. The data consists of a set of values #k defined at the locations # ### #####. A value at a point # # ## k ## k## # can be interpolated using a Hermite interpolant as follows [8] ####### ## # # k # # # ## ## # # k # # # ## ## # # k #### # where ## # #k # #k## ##k ## # ##k # ##k # #k## ## # #k ## # #k and #k ###k## # #k # #### #k # #k## # #k # However, this interpolant usually overshoots the data as we show on the left hand side of Figure 9. We want to avoid ....

J. D. Fowley, A. van Dam, S. K. Feiner, and J. F. Hughes. Computer Graphics: Principles and Practice. Second Edition. Addison-Wesley, Reading, MA, 1990.

....in turn, it should suffice to break them down into the three categories relevant to this project: projection, modeling, and rasterization. 2.1. 1 Projection In general, projections transform points in a coordinate system of dimension n into points in a coordinate system of dimension less than n [3]. More specifically for OpenGL, the projection matrix transforms coordinates in 3D space into coordinates on a 2D plane suitable for viewing on a computer screen. The main two methods of performing this transformation are parallel and perspective. The easiest way to explain the difference in the ....

Feiner, Foley, Hughes, van Dam. Computer Graphics: Principles and Practice. Addison-Wesley, 1997.

....5.1 Hardware Based Renderer In our implementation of the hardware based renderer, we follow the algorithm outlined in [17] In a first pass, we compute the amount of light that directly reaches each voxel of the grid. This is achieved using a fast Bresenham line drawing voxel traversal algorithm [8]. Initially the transparencies of each ray are set to one ( 4 C : Then, each time a voxel is hit, the transparency is computed from the voxel s density: 4 9 # , where is the grid spacing. Then the voxel s radiance is set to ( 9 ( ....

....along each coordinate axis. Therefore, it is sufficient to describe the one dimensional case only. The data consists of a set of values defined at the locations 3 . A value at a point HI5 can be interpolated using a Hermite interpolant as follows [8] 5 D where D HI5 # # D HI5 5 D and D HI5 S365 F # HI5 However, this interpolant usually overshoots the ....

J. D. Fowley, A. van Dam, S. K. Feiner, and J. F. Hughes. Computer Graphics: Principles and Practice. Second Edition. Addison-Wesley, Reading, MA, 1990.

....5.1 Hardware Based Renderer In our implementation of the hardware based renderer, we follow the algorithm outlined in [17] In a first pass, we compute the amount of light that directly reaches each voxel of the grid. This is achieved using a fast Bresenham line drawing voxel traversal algorithm [8]. Initially the transparencies of each ray are set to one (Tray = 1) Then, each time a voxel is hit, the transparency is computed from the voxel s density: Tvox = exp( GammaC ext h) where h is the grid spacing. Then the voxel s radiance is set to Lvox = Omega L light (1 Gamma Tvox ) Tray ; ....

....along each coordinate axis. Therefore, it is sufficient to describe the one dimensional case only. The data consists of a set of values fk defined at the locations k = 0; Delta Delta Delta ; N . A value at a point t 2 [t k ; t k 1 ] can be interpolated using a Hermite interpolant as follows [8] f(t) a3 (t Gamma t k ) 3 a2 (t Gamma t k ) 2 a1 (t Gamma t k ) a0 ; where a3 = dk dk 1 Gamma Delta k a2 = 3 Delta k Gamma 2dk Gamma dk 1 a1 = dk a0 = fk and dk = fk 1 Gamma fk Gamma1 ) 2; Delta k = fk 1 Gamma fk : However, this interpolant usually overshoots the ....

J. D. Fowley, A. van Dam, S. K. Feiner, and J. F. Hughes. Computer Graphics: Principles and Practice. Second Edition. Addison-Wesley, Reading, MA, 1990.

....(e.g. 10, 7] corresponds to a data partitioning, where each node is assigned a subset of the polygons in the scene without any restrictions on the position of the polygons. Each frame, once each node has rendered its assigned polygons, the pixels must be sorted, typically using Z buffering [5]. While compatible with our environment, sort last is less than ideal because: i) rendering nodes must send their Z buffers along with the rendered pixels for the composition of the final image, which approximately doubles the required bandwidth, and (ii) primitives are typically assigned to ....

....sub polygons if it intersects more than one leaf octant. Figure 2(a) shows an example partition albeit of a 2D area for ease of presentation. Given the leaf octants of an octree spatial decomposition, we use a simple visibility property of rectangular volumes to construct the occlusion graph [5, 13] (Figure 2(b) We estimate the rendering time of each aggregate scene object in frame f by measuring the object s rendering time in frame f 1. We estimate the footprint of each object by projecting the bounding box of the polygons in it onto a course 2D grid representing the viewport as shown in ....

J. D. Foley, A. van Dam, S. K. Feiner, and J. F. Hughes. Computer Graphics: Principles and Practice. AddisonWesley, Reading, MA, 1990.

....only organic forms, i.e. shapes with rounded edges and corners. Additional representations of shape for computer graphics include look up tables, Fourier expansions, particle systems, grammar based models, and fractals (iterated function systems) all of which tend to have focused applications [10]. The ADF representation, its applications, and the implementation details presented in this paper are new. Sampled distance fields have, however, been used previously in a number of specific applications. They have been used in robotics for path planning [12, 13] and to generate swept volumes ....

J. Foley, A. van Dam, S. Feiner, and J. Hughes, Computer Graphics: Principles and Practice, Addison-Wesley, 1992.

....the angle between the viewing direction and the specular direction, and n is the specularity of the material, the computational cost grows at least logarithmically in n. 2 Background and Related Works 2.1 3D Graphics Rendering System A typical 3D computer graphics pipeline is depicted in Fig. 1 [13]. The rst step, the application, depends on the host CPU computing power. The second step, he geometric engine, does intensive oating point arithmetic, thus it needs multiple pipelined oating point arithmetic units and or speci c arithmetic function units, such as a oating point 4 4 matrix ....

J. Foley, A. van Dam, S. Feiner, and J. Hughes. Computer Graphics: Principles and Practice. Addison-Wesley Publishing Company, second edition, 1990.

.... that by representing virtual objects in an affine reference frame and by performing computer graphics operations such as projection and visible surface determination directly on affine models, the entire video overlay process is described by a single 4 Theta4 homogeneous view transformation matrix [30]. Furthermore, the elements of this matrix are simply the image x and y coordinates of fiducial points. This not only enables the efficient estimation of the view transformation matrix but also leads to the use of optimal estimators such as the Kalman filter [31 33] to track the fiducial points ....

....affine object models from live video images. Limitations of our approach are summarized in Section 8. 2 Geometrical Foundations Accurate projection of a virtual object requires knowing precisely the combined effect of the object toworld, world to camera and camera to image transformations [30]. In homogeneous coordinates this 5 projection is described by the equation 2 6 6 6 4 u v h 3 7 7 7 5 = P 3 Theta4 C 4 Theta4 O 4 Theta4 2 6 6 6 6 6 6 4 x y z w 3 7 7 7 7 7 7 5 (1) where [x y z w] T is a point on the virtual object, u v h] T is its ....

[Article contains additional citation context not shown here]

J. D. Foley, A. van Dam, S. K. Feiner, and J. F. Hughes, Computer Graphics Principles and Practice. Addison-Wesley Publishing Co., 1990.

....is often called alpha blending [8] However, alpha blending has two major disadvantages: The transparent polygons have to rendered in a back to front order and the polygons are not allowed to intersect. A technique that does not suffer from these drawbacks is so called screen door transparency [2]. Screendoor transparency relies on the capability of the human eye to perform spatial integration on the final scene to simulate nonrefractive transparency. A polygon s opacity is now expressed with a 2D pixel mask. This pixel mask determines how many and which pixels of a projected polygon are ....

J. Foley, A. van Dam, S. Feiner, and J. Hughes. Computer Graphics: Principles and Practice. AddisonWesley, second edition, 1990.

.... of the surface over which the visible rim slides during the generated motions (i.e. it marks the portions of the object that would be reconstructed due to these motions) At each viewpoint, the object s visible rim and its occluding contour are computed using efficient ray tracing techniques 6 [46]. Below, we briefly present results of using this system to simulate the reconstruction of the surface of the curved pipe shown in Figure 1. These results can be evaluated in three ways: 1) The surface regions reconstructed, 2) the number of times the Region Reconstruction Strategies were ....

J. D. Foley, A. van Dam, S. K. Feiner, and J. F. Hughes, Computer Graphics Principles and Practice. Addison-Wesley Publishing Co., 1990. 37

....pixels in each image and voxels in the plenoptic decomposition. In particular, we need to compute (1) for each voxel which pixels it projects to in each image, and (2) for each pixel, what is the closest voxel that projects to that pixel. Both tasks are accomplished using an item buffer technique[40], in which each voxel is assigned a unique pseudo color and rendered to each viewpoint, using Z buffering to account for visibility. The resulting images provide a direct two way mapping between voxels and pixels in every view. Our implementation used this technique to allow a user to paint in ....

....for viewing by the user and warped automatically by the system in (d) by propagating the user specified warp in (a) simulated by evaluating Eq. 1) with the desired rotation matrix M. A variety of natural user interfaces exist for specifying rotations via direct manipulation, see for example [40]. 23 5.2 Rendering Voxel Models Once the illumination is set and voxel radiances are determined, a new image is generated by assigning colors (radiances) to the voxels and projecting the model to the desired viewpoint. This projection can be computed very efficiently using voxel splatting ....

J. D. Foley, A. van Dam, S. K. Feiner, and J. F. Hughes, Computer Graphics Principles and Practice. Addison-Wesley Publishing Co., 1990.

....Some GUI objects are just a wrapping around existing widgets implemented in TclTk, but others hide a complex hierarchy of events and widgets, such as the graphic input tools or the visualization windows. The new GUI objects described below are based on concepts borrowed from Computer Graphics [10]: 1. Visualization Window: high level widget to display images and graphics (curves) with local basic viewing control (see figure 1) Components are: ffl one basic image displayed in grey scale; ffl mask images that define translucent color layers combined with the basic image (only primary ....

J.D. FOLEY, A. VAN DAM, S.K. FEINER, and J.F. HUGHES. Computer Graphics - Principles and Practice. Addison-Wesley, 1990.

....color space The hue (H) saturation (S) and value (V) system is based on a warped version of the RGB space and is directly related to intuitive color notions of hue, saturation and brightness. The conversion from the RGB color space to the HSV color space is done through the following equation [1]: h 0 = 8 : g Gammab) ffi if r = max 2 (b Gammar) ffi if g = max 4 (r Gammag) ffi if b = max (6) v = max s = max Gamma min max h = h 0 60; 7) where max = MAX(r; g; b) min = MIN(r; g; b) ffi = max Gamma min, and (h; s; v) is the corresponding point of (r; g; b) ....

.... the distance from color C 1 = X 1 ; Y 1 ; Z 1 ) to color C 1 DeltaC , and the distance from color C 2 = X 2 ; Y 2 ; Z 2 ) to color C 2 DeltaC , where DeltaC = DeltaX; DeltaY; DeltaZ ) Both distances are equal to DeltaC , yet in general they will not be perceived as being equal [1]. Perceptual uniformity means that the same DeltaC at two different points in the color space makes the equal perceivable color difference. CIE XYZ, RGB, and HSV color spaces do not exhibit perceptual uniformity. There are two perceptually uniform color spaces that are the agreed standards in ....

J. D. Foley, A. van Dam, S. K. Feiner, and J. F. Hughes. Computer Graphics: Principles and Practice. AddisonWesley Publishing, 1996.

....face pairs of two polyhedra with prescribed face counts when the polyhedra are convex, and lower bounds in the general case. 1 Introduction Computing the intersection of two polygons is a fundamental problem arising in computational geometry and related fields such as computer graphics [1]. In analyzing the complexity of algorithms to solve the problem, the following question arises: what is the maximum number of intersections between two polygons, given the number of edges of each Approximate analysis is easy: if two polygons have p and q edges, then in the worst case there can ....

J. D. Foley, A. van Dam, S. K. Feiner, and J. F. Hughes. Computer Graphics: Principles and Practice. Addison-Wesley, Reading, MA, second edition, 1990.

....is 80 60, 24 bit color. The 326 images were randomly divided into two halves, with 163 images as training data and the other 163 images as test data. Figure 2 shows random samples of the training data set. Figure 2. Some images in the training data set In this work we use the HSV color space [11] instead of the RGB color space. Moreover, we use only Hue and Saturation and ignore V (brightness) in order to minimize the influence of shadow and uneven lighting. We plot HueSaturation histograms for analysis. Since the images are manually segmented, we are able to plot the color histograms for ....

J. D. Foley, A. van Dam, S. K. Feiner, J. F. Hughes. Computer Graphics: Principles and Practice. 2nd ed. p. 590. Addison-Wesley, Mass., 1993

....et al. Artificial Intelligence 105 (1998) 47 75 63 This can be done in O(N) steps, where N is the number of vertices in the scene. 3) All the hidden edges can be removed by one of the many existing techniques in computer graphics and computational geometry. We have used Appel s algorithm [1,6], which suits our case well. The algorithm has complexity O(N log N) The overall complexity of the algorithm for obtaining the line drawings from the scene is therefore O(N log N) whereN is the number of vertices in the scene. In terms of the number n of generating planes, the complexity is O(n ....

J.D. Foley, A. van Dam, S.K. Feiner, J.F. Hughes, Computer Graphics: Principles and Practice, AddisonWesley, Reading, MA, 1990.

....the achievable frame rate. An alternative is to use a shading approximation by pairwise subtracting coplanar texture slices, one shifted in direction of the light source [PAC97] System Architecture Texture mapping hardware is part of the raster graphics rendering pipeline shown in Figure 5 [FvDFH90] During planar texture resampling, the geometry processor Geometry Processor Rasterizer Frame Buffer Display Texture Memory Texture Engine Fig. 5. Texture mapping system architecture. computes the texture coordinates of each vertex of a resampling slice. The rasterizer is connected to ....

J. Foley, A. van Dam, S. Feiner, and J. Hughes. Computer Graphics: Principles and Practice. Addison-Wesley Publishing Company, 2nd edition, 1990.

....removal was proposed by Georges [70] and Franklin [62] Further in the text, we restrict our discussion to the most popular algorithms nowadays and practical algorithms that aim to achieve the output sensitivity or to increase the amount of coherence exploited. The commonly used depth buffer [60] (or z buffer) is a typical discrete VFP algorithm. It solves the visibility for each pixel of the screen by comparing depth values of objects that project to the pixel. The depth values can be obtained easily during the object rasterization. In a way it resembles the ray casting algorithm, that ....

....a topological information about a polygonal scene in a binary tree. The tree is build in the object space and is used to obtain a depth ordering of the polygons with respect to any point in the scene. The polygons are processed in back to front order and the discrete painter s algorithm [60] is applied to resolve pixel visibility by overdrawing its previous contents. The depth buffer and the BSP assisted painter s algorithm are not output sensitive since they must process (rasterize) every object in the scene y . Thus they may spend significant Many objects are pierced by each ....

J. D. Foley, A. van Dam, S. K. Feiner, and J. F. Hughes. Computer Graphics: Principles and Practice. Addison-Wesley Publishing Co., Reading, MA, 2nd edition, 1990. T 385.C587. 49

....segmented. Each image is 80 # 60, 24 bit color. The 326 images were randomly divided into two halves, with 163 images as training data and the other 163 images as test data. Figure 2 shows random samples of the training data set. 2. in the data set In this work we use the HSV color space [11] instead of the RGB color space. Moreover, we use only Hue and Saturation and ignore V (brightness) in order to minimize the influence of shadow and uneven lighting. We plot HueSaturation histograms for analysis. Since the images are manually segmented, weareable to plot the color histograms for ....

J. D. Foley, A. van Dam, S. K. Feiner, J. F. Hughes. Computer Graphics: Principles and Practice. 2nd ed. p. 590. Addison-Wesley, Mass., 1993

....will concentrate particularly on facial model and animation. Then Section 8.3.3 will discuss the problem of lip shape computation during speech. Finally Section 8.4 introduces what behaviors and functions are necessary to create a conversational agent. More detailed information can be found in [283, 140, 304]. ffl Model creation: Different steps are required to create a facial model: acquisition of the model, geometric representation, color determination. Various techniques are available to generate a model, and more specifically a facial model. One can use a modeler, one can acquire data from ....

....effects. Shading an object requires to select the number of lights and the type of lights illuminating the scene. Shadow is important in a scene. It gives it a 3D look and can produce dramatic effects. Lighting computation can be extremely complex. The simplest illumination models are presented in [140, 283]. The picture of a face can be texture mapped onto the facial model. It gives good result for the eyes and is a simple way to model the teeth and the inside of the mouth. The matching between the facial features of the texture and of the model requires special attention. Other techniques producing ....

J.D. Foley, A. van Dam, S.K. Feiner, and J.F. Hughes. Computer Graphics: Principles and Practice. Addison-Wesley, Reading, MA, 1990. Second Edition.

....doing so. A straightforward approach models the three RGB values by the line spectrum consisting of the corresponding primaries[9] Sun et al. 26] propose the representation by Gaussian functions with adapted width. Another approach uses Fourier functions[9] We find the dominant wavelength model[7] useful because it provides a smooth change of color and brightness for a wide range of Doppler factors. The corresponding spectral power distribution consists of a spike at the dominant wavelength and of a uniform distribution, i.e. white light. The luminance and excitation purity determine the ....

....distribution consists of a spike at the dominant wavelength and of a uniform distribution, i.e. white light. The luminance and excitation purity determine the levels of the two parts of the spectrum. The parameters for the dominant wavelength model can be computed from RGB values according to [7]. The relativistic situation requires only one slight extension of the original model. Here, the uniform part of the spectrum is not restricted to the range of visible wavelengths, but comprises a larger interval. In this way, the spectrum is still present after Doppler shifting. With the ....

J. D. Foley, A. van Dam, S. K. Feiner, and J. F. Hughes. Computer Graphics: Principles and Practice. Addison-Wesley, Reading, Massachusetts, 1990.

....In the following discussion, we will concentrate on secondorder B splines. However, this theory can be generalized to higher order curves. Also, it should be noted that first order B splines are equivalent to polygons. A B spline is a specific curve type from the family of parametric curves [53]. A parametric curve consists of one or more curve segments. Each curve segment is defined by control points, where defines the degree of the curve. The control points are located around the curve segment, and together with a constant base matrix , the control points solely define the shape of the ....

....the entire curve , consisting of curve segments, which is of the following form: 9) Among common parametric curves are the Bezier curve and the B spline curve. For the proposed shape coding method in this section, we chose a second order (quadratic) basis uniform nonrational B spline curve [53] with the following base matrix: 10) Fig. 24 shows such a second order B spline curve. The shape coding method presented in this section is independent of the matrix and degree , that is, parametric curves of higher order can be used. 1) Double Control Points: The beginning and the end of the ....

J. D. Foley, A. van Dam, S. Feiner, and J. Hughes, Computer Graphics: Principles and Practice. Reading, MA: AddisonWesley, 1990, pp. 478--516.

....on soft tissue. inormalj time iaccelaratedj time linear elasticity 15000 ms 7 ms quasi non linear elasticity 50000 ms 8 ms Table 1. Computing times for a liver model. 5 Collision detection Collision detection is a complex and well known problem in computer animation [40] 41] 42] [43], 44] 45] When the real time constraint is added, the diOEculty is considerably increased. With physically based models, most of the external forces are contact forces and in surgery simulation, the deformation is mainly driven by user interactions so an eOEcient collision detection algorithm ....

J.D. Foley, A. Van Dam, S. Feiner, and J. Hughes, Computer Graphics - Principles and Practice, Addison Wesley, 1990.

No context found.

Feiner, S.K., Foley, J. D., Hughes, J.F., van Dan, A. Computer Graphics: Principles and Practice. 2nd Edition, Addison Wesley, 1992.

No context found.

J. D. Foley, A. van Dam, S. K. Feiner, and J. F. Hughes. Computer Graphics --- Principles and Practice. AddisonWesley, second edition, 1993.

First 50 documents

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC