### Table 2. Performance with image warping

2001

"... In PAGE 9: ... Figure 10 displays a couple of scenes from our virtual colonoscopy system. Table 1 and Table2 shows the performance of our system. Without reusing slab images, although only a small portion of voxels is picked by the slab structure and VolumePro is exploited, the system only delivers half frame per second.... ..."

Cited by 6

### Table 6: A framework for gradient descent image alignment algorithms. Gradient descent image alignment algorithms can be either additive or compositional, and either forwards or inverse. The inverse algorithms are computationally efficient whereas the forwards algorithms are not. The various algorithms can be applied to different sets of warps. Most sets of warps in computer vision form groups and so the forwards additive, the forwards compositional, and the inverse compositional algorithms can be applied to most sets of warps. The inverse additive algorithm can only be applied to a very small class of warps, mostly linear 2D warps. Algorithm For Example Efficient? Can be Applied To

2004

"... In PAGE 25: ...teration to first order in A1D4. In Section 3.4 we validated this equivalence empirically. The four algorithms do differ, however, in two other respects. See Table6 for a summary. Although the computational requirements of the two forwards algorithms are almost identical and... ..."

Cited by 144

### Table 6: A framework for gradient descent image alignment algorithms. Gradient descent image alignment algorithms can be either additive or compositional, and either forwards or inverse. The inverse algorithms are computationally efficient whereas the forwards algorithms are not. The various algorithms can be applied to different sets of warps. Most sets of warps in computer vision form groups and so the forwards additive, the forwards compositional, and the inverse compositional algorithms can be applied to most sets of warps. The inverse additive algorithm can only be applied to a very small class of warps, mostly linear 2D warps. Algorithm For Example Efficient? Can be Applied To

2004

"... In PAGE 25: ...teration to first order in A1D4. In Section 3.4 we validated this equivalence empirically. The four algorithms do differ, however, in two other respects. See Table6 for a summary. Although the computational requirements of the two forwards algorithms are almost identical and... ..."

Cited by 144

### Table 6: A framework for gradient descent image alignment algorithms. Gradient descent image alignment algorithms can be either additive or compositional, and either forwards or inverse. The inverse algorithms are computationally efficient whereas the forwards algorithms are not. The various algorithms can be applied to different sets of warps. Most sets of warps in computer vision form groups and so the forwards additive, the forwards compositional, and the inverse compositional algorithms can be applied to most sets of warps. The inverse additive algorithm can only be applied to a very small class of warps, mostly linear 2D warps. Algorithm For Example Complexity Can be Applied To

2004

"... In PAGE 30: ... In Section 3.4 we validated this equivalence empirically. The four algorithms do differ, however, in two other respects. See Table6 for a summary. Although the computational requirements of the two forwards algorithms are almost identical and the computational requirements of the two inverse algorithms are also almost identical, the two inverse algorithms are far more efficient than the two forwards algorithms.... In PAGE 49: ...pproximations, and the Levenberg-Marquardt approximation. These two choices are orthogonal. For example, one could derive a forwards compositional steepest descent algorithm. The results of the first half are summarized in Table6 . All four algorithms empirically perform equivalently.... ..."

Cited by 144

### TABLE I. Smoothness measurements of the averaged warped images

### Table 2: Timings for noninteractive warping

"... In PAGE 53: ...Table 2: Timings for noninteractive warping The image calculated for Table2 contains only translational warps. Scal- ing warps are somewhat slower because their mapping function includes a square root, and rotation warps are the slowest because their mapping function requires the calculation of a sine/cosine pair.... ..."

### Table 3: Software vs. texture based warp. Warp type Output image size PC [ms] Onyx2 [ms]

### Table 1. Time Warp: Sensors and Indicators

"... In PAGE 61: ... It is possible to omit certain characteristics if they are irrelevant for the particular system under test or if they do not provide any information useful in the performance evaluation study. Table1 summarizes the characterization of the three layers. Descriptions Layer Functional Sequential Parallel Quantitative application structural multiplicity Application set of algorithms behavior parallelism of algorithms, graph graph volume of data algorithm structural multiplicity Algorithm set of routines behavior parallelism of routines, graph graph volume of data routine structural multiplicity Routine set of statements behavior parallelism of statements, graph graph volume of data Table 1.... In PAGE 61: ... Table 1 summarizes the characterization of the three layers. Descriptions Layer Functional Sequential Parallel Quantitative application structural multiplicity Application set of algorithms behavior parallelism of algorithms, graph graph volume of data algorithm structural multiplicity Algorithm set of routines behavior parallelism of routines, graph graph volume of data routine structural multiplicity Routine set of statements behavior parallelism of statements, graph graph volume of data Table1 . Characterization at the three hierarchical layers Functional Description The identi cation of the basic components for each... In PAGE 69: ...Action Netscape start, quit Windows open, close Conversations http, ftp Methods GET Table1 : Types of Actions at eachLevel web browsers (level 2). The time between starting a browser and quitting it is referred to as the life-time of the browser and is represented at the second level byathick line.... In PAGE 70: ... Thus, we are only considering the 4 inner levels in the hierarchy. Table1 summarizes these four levels and the actions that could occur at each of them. Applying the proposed approachofPACFG for workload modeling in performance evaluation studies has already been demonstrated in previous papers [Ragh 93, SV R 94, Ragh 95, SV R 96].... In PAGE 86: ...246 seconds. The MIME types of the downloaded file were grouped together, forming the five groups stated in Table1 . Each session then was represented by its mixture of downloaded filetypes in percent, yielding 196 5-dimensional vectors.... In PAGE 86: ... CC BD CC BE CC BF CC BG CC BH appl. audio/video image text unknown Table1 . MIME type groups.... In PAGE 98: ... In this work we study the distributed execution of Time Warp logical processes (LPs) under the performance super- vision of the PPMM environment (Figure 3). State and performance characteristics of a Time Warp LP are sampled at runtime via the sensors listed in Table1 . Time Warp is knownrequire a sufficient amount ofmemory to be operable.... In PAGE 99: ... The structure of the simulation model guarantees balanced load among the four LPs, and the communication structure among LPs adheres to a ring topology. The eight sensors outlined in Table1 have been inserted into the Time Warp code. The point sample sensor (PS) as well as each of the cumulative sensors (CS) are transformed into a corresponding indicators (see Table 1).... ..."

### Table 3: Number of arithmetic operations in different types of warps. The columns labeled planar, cylindrical, and spherical refer to an efficient incremental implementation of the standard warp in each of these three cases.

Cited by 3

### Table 3: Number of arithmetic operations in different types of warps. The columns labeled planar, cylindrical, and spherical refer to an efficient incremental implementation of the standard warp in each of these three cases.

2001