Richard Szeliski and David Tonnesen. Surface modeling with oriented particle systems. Computer Graphics (SIGGRAPH '92 Proceedings), 26(2):185--194, July 1992.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....circular splats to render points, they limited the minimum splat size to a pixel size to avoid aliasing. They used accumulated weights for coverage estimation and an a buffer [10] for visibility and composition in the same way as Zwicker et al. 67] 16 years later. In 1992 Szeliski and Tonneson [56] used oriented particles for surface modeling and interactive editing. The oriented particles were points with local coordinate frame, interacting with each other by long range repulsion forces and short range attraction forces. For visualization of oriented particles, however, they did not use ....

....final production they constructed a surface triangulation. They considered the oriented particles as surface elements, surfels . In 1994 Witkin and Heckbert [61] used oriented particles for sampling and interactive editing of implicit surfaces. The particles were visualized in the same way as in [56]. In 1998 point rendering was revisited by Grossman and Dally [25, 26] This time the aim was to develop an output sensitive rendering algorithm for complex objects that would support dynamic lighting. The work was mainly inspired by advances in image based rendering. A real boom of point based ....

[Article contains additional citation context not shown here]

Richard Szeliski and David Tonnesen. Surface modeling with oriented particle systems. In Siggraph '92 Proceedings, pages 185--194, 1992. 3, 6

.... computation, # availability of information about contact between particles, # availability of orientation information from neighbors, and # particle control via conditional differential equations 137 Siggraph 97 Course Notes 138 Some of these are available in other recent particle systems [15, 16, 10, 18, 20, 19, 22]. The remainder of this paper is a modified version from Chapter 2 of my thesis [3] and describes the simulator from the biological standpoint with which it was constructed. I hope this will clarify some of the design decisions that we could not thoroughly discuss in the limited length of a ....

Richard Szeliski and David Tonnesen. Surface modeling with oriented particle systems. In Edwin E. Catmull, editor, Computer Graphics (SIGGRAPH '92 Proceedings) , volume 26, pages 185--194, July 1992.

....and how the reference pictures that encode the attributes are created. Finally, we present various ways of manipulating the brush stroke attributes to produce painterly images. 3. 1 Generating Particles There are many methods of populating a surface with particles, such as those described in [15] and [18] We employ a simple method that starts with a parametric surface and a desired number of particles. We tessellate the surface into triangles that approximate the surface. Then, for each triangle, we compute its surface area and randomly distribute particles within it. The number of ....

Richard Szeliski and David Tonnesen. Surface modeling with oriented particle systems. In ComputerGraphics (SIGGRAPH '92 Proceedings), volume 26, pages 185--194, July 1992.

....Further, no real consideration has been given to the problem of creating nontrivial smooth surface topologies interactively. The one approach to fair shape design that has allowed large scale changes in shape and topology during sculpting is the oriented particle system of Szeliski and Tonnesin[34]. The drawback of this approach is that there is no explicit control over surface topology because there is no actual surface. A surface triangulation can be imposed on the particles strictly as an output, but this has no influence on the particles subsequent behavior, and no persistence ....

Richard Szeliski and David Tonnesen. Surface modeling with oriented particle systems. Computer Graphics, 26(2), July 1992. (Proceedings Siggraph '92).

....point on the medial axis of the surface [1] Our rendering approach assumes that the maximum size of a gap in the samples is known. If the point samples are not that uniformly distributed over the surface, they can be made to be more evenly spaced using an approach described by Szeliski et al. [21, 22]. By giving attracting and repulsive forces to the points, points tend to even out the spacing between them. Adding points at gaps along borders can fill holes of a given maximum size. Like Szeliski et al. we also assume a normal to be available per point. If the 3D scanning process does not ....

Richard Szeliski and David Tonnesen: "Surface modeling with oriented particle systems", Proc. SIGGRAPH '92, pp 185-194, 1992.

....et al. (1993) and in a related way by Galyean [Par77, Nay90, GH91, MFD93] Mingxian uses CSG based virtual modeling tools in a non free form CAD system. Pushing and pulling points as we propose to do has been done to some extent by Brewer (1977) Hsu (1992) and Szeliski (1992) BA77, HHK92, ST92] Brewer pushed parametric surfaces control points with a planar tool. Szeliski used tools to apply and cancel forces in a particle based modeler. Hsu used pushing and pulling tools to impart free form deformation (FFD) his work is described in 10 Section 2.3 on alternate approaches to ....

....technique or bounding box tests could be used to generate a set of the vertices in the general neighborhood of the tool. There exist many methods to do this, which vary greatly in complexity ( Sam90] Previous researchers in interactive modeling have chosen BSP trees ( Nay90] and B trees ( ST92] It seems that the grid file method, which is a sort of 3D hashing scheme described by Samet [Sam90] is likely to be effective due to its simple method of hashing a point in 3D space to one or more hash buckets in which the vertices near the given point are listed. One major current problem ....

Richard Szelski and David Tonnesen. Surface modeling with oriented particle systems. Computer Graphics (ACM SIGGRAPH Proceedings), 26(2):185--194, July 1992.

....and linear triangular elements are used. Comparing (2.9) to (2.10) we see that in the latter, the stiffness matrix K does not depend on q as in (2.9) However, the term g q is introduced in (2. 10) Recent work with deformable models and surfaces employs a technique based on particle systems [43, 50]. The interaction between particles can be modeled in different ways. In [50] CHAPTER 2. DYNAMIC SIMULATION OF FLEXIBLE MODELS 20 concepts from molecular dynamics are used in order to model fluids using thermal particles. Potential functions between pairs of particles are defined as a function of ....

....SIMULATION OF FLEXIBLE MODELS 20 concepts from molecular dynamics are used in order to model fluids using thermal particles. Potential functions between pairs of particles are defined as a function of the temperature in order to establish the required level of rigidity of the simulated fluid. In [43] surface particles are used to model deformable surfaces. The particles have an orientation which is dynamically constrained according to the desired smoothness of the surface. Our formulation with respect to dynamics is most closely related to the one presented in [45] Our equations of motion ....

Richard Szeliski and David Tonnesen. Surface modeling with oriented particle systems. Computer Graphics, 26(2):185--194, July 1992.

....is a parameter to the reconstruction procedure. Fish scales can then be ltered based on the goodness of t and on their rank. See [4] for more details. All the processes described above introduce Figure 3: Fish scales are disk like oriented structures much like the oriented particles of Tonnesen [5] (left) In our actual implementation, a sh scale is a fuzzy set with an ellipsoidal kernel (right) given by its center x 0 and structural matrix S. various artifacts and biases to the set of reconstructed sh scales. Their position in space is accurate enough for creating a triangulated ....

Richard Szeliski and David Tonnesen. Surface modeling with oriented particle systems. Computer Graphics (SIGGRAPH '92), 26(2):185-194, July 1992.

....reprezentations: volume based and surface based. Volume based models (e.g. general cylinders, quadrics and CSG) strongly generalize the object shape. That is why the surface based models (e.g. triangulations [dFP85] based on k simplex meshes [Del94] alpha shapes [Ede92] oriented particles [LP95, ST92] are used very much today. 3D reconstruction is difficult, because: In this time R. ##ra works in GRASP Lab, University of Pennsylvania. ffl Real scene often consists several objects with varied details. These objects can contain edges, ovals or holes. ffl Input data can be obtained from ....

Richard Szeliski and David Tonnesen. Surface modeling with oriented particle systems. Computer Graphics (SIGGRAPH '92), 26(2):185--194, July 1992.

....even triangulations, avoiding skinny triangles like those in (a) that lead to unexpected wiggles and bumps when subdivided. gorithm treats vertices as particles that interact with their neighbors while being guided by an implicit function, it resembles the particle system methods described in [21, 25]. An important difference is that skin particles have explicit connectivity information. Skin is related to the large body of work on implicit surfaces, particularly offset and convolution surfaces [3, 17] and blobby modeling [2, 4] in the constructive approach to modeling it provides. For ....

Richard Szeliski and David Tonnesen. Surface modeling with oriented particle systems. In SIGGRAPH 92 ConferenceProceedings,pp. 185-- 194. ACM SIGGRAPH, July 1992.

....be other applications. Recent work on manipulation of models makes use of non physical motion [6] 15] Finally, it is important to note the difference between positionbased physics and particle based systems. Examples of particlebased systems are too numerous to note all of them. See [32] 17] [31] [33] 10] 9] 21] for recent work. In general, particlebased systems model moving particles with forces between them, not rigid colliding objects. Simulating rigid objects using particles requires a very steep repulsive energy gradient, and hence particle based systems are subject to the same ....

Richard Szeliski and David Tonnesen. Surface modeling with oriented particle systems. In Edwin E. Catmull, editor, Computer Graphics (SIGGRAPH '92 Proceedings), volume 26, pages 185--194, July 1992.

....explored mesh simplification, framing it as an optimization problem with penalties for geometric error, number of samples, and edge length [16] Their method did not restrict the points to a surface, however, as Turk s and ours do. Szeliski and Tonnesen used oriented particles to model surfaces [27]. Their technique allowed the user to move the particles interactively, employing short range repulsion and long range attraction to keep the particles from clumping or flying apart. The system generated a surface by connecting neighboring particles appropriately, but it did not manipulate a high ....

....than finding a polygonization from a set of samples on a grid in 3 D, as in marching cubes algorithms, where an approximate topology is suggested by the signs of the samples and by the topology of the grid itself. Delaunay triangulation in 2 D or 3 D is one possible way to extract topology [12,27]. A more robust alternative would employ Lipschitz conditions and interval arithmetic [26] To preserve the basic advantages of our method, we would require a polygonization algorithm that allows efficient dynamic updates as the surface changes. Although we developed it to sample implicit ....

Richard Szeliski and David Tonnesen. Surface modeling with oriented particle systems. Computer Graphics (SIGGRAPH '92 Proceedings), 26(2):185--194, July 1992.

....et al. (1993) and in a related way by Galyean [Par77, Nay90, GH91, MFD93] Mingxian uses CSG based virtual modeling tools in a non free form CAD system. Pushing and pulling points as we propose to do has been done to some extent by Brewer (1977) Hsu (1992) and Szeliski (1992) BA77, HHK92, ST92] Brewer pushed parametric surfaces control points with a planar tool. Szeliski used tools to apply and cancel forces in a particle based modeler. Hsu used pushing and pulling tools to impart free form deformation (FFD) The most relevant work is that of Naylor and Galyean; this will be ....

Richard Szelski and David Tonnesen. Surface modeling with oriented particle systems. Computer Graphics (ACM SIGGRAPH Proceedings), 26(2):185--194, July 1992.

No context found.

Richard Szeliski and David Tonnesen. Surface modeling with oriented particle systems. Computer Graphics (SIGGRAPH '92 Proceedings), 26(2):185--194, July 1992.

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