Coherence in Volume Rendering Forschungsseminar WS 98/99 Institut für Computergrafik

Anna Vilanova and Katja Bühler

Complete notes

(Some images might got lost during the LaTex->HTML conversion)
of our research and the talk are also available in postscript format :  FoSe_notes.ps.gz(12MB)

Articles on Coherence in Volume Rendering

• General

[Yagel '96],[Groeller, Purgathofer '95]
• Pixel Space Coherence (Katja)

[Bergman et al. '86],[Levoy '90b]
• Object Space Coherence (Katja)

[Danskin, Hanrahan '92] extension of [Laur, Hanrahan '91](Space Leaping) and [Levoy '90a](Splatting) All Algorithms are based on preprocessing the data storing them in octrees together with certain information about the homogenity of the volumetric area corresponding to each node of the tree.
Proximity Clouds: [Yagel, Shi '93].
• Frame-to-Frame Coherence (Anna)

[Brady et al. '97],[Brady et al. '98],[Yagel, Shi '93],[Gudmundsson, Randen '90]
• Inter Ray Coherence/Shear-Wrap (Anna)

[Lacroute, Levoy '94],[Yagel, Kaufman '92]
• Stereoscopic Images (Anna + Katja)

[Adelson, Hansen '94],[He, Kaufman '96a]
• Wavelets (Katja)

[Sánchez, Carvajal '98]
• others

[Avila et al. '94],[Cohen-Or, Fleishman '95],[Mueller, Yagel '96](Splats, no coherence)
 

References

Talk structure (Katja's part)

1. Title + "philosophical" definition
2. Definitions of the term "coherence"
1. Dictionaries
2. "Meister"
3. Motivation
4. Definitions: Pixel + object space coherence, ray coherence + frame-to-frame coherence
5. Exploiting pixel space coherence: Adaptive raycasting.
1. sentence "it is highly probable...."
2. the algorithm
3. problems: the selection of omega_min, omega_max, epsilon
4. results
6. Exploiting object space coherence: Adaptive refinement.
1. Central data structures: Pyramides and octrees. Subdividing the space in an adaptive manner with respect to a certain area coherence (empty/non-empty areas, homogeneous areas).
2. Preprocessing allows fast traversal of empty areas (space leaping) and fast sampling of homogeneous areas.
7. Exploiting object space coherence: Proximity Clouds

Algorithm + Advantages with respect to 6.
8. Detecting areas with constant linear and non-linear behaviour via Wavelets.
9. Coherence in stereoscopic visualization.
1. What is stereoscopic visualization, the three steps
2. Quality mesurement
3. The viewing transformation
4. The reprojection
5. Segment decomposition
6. The early ray termination problem
7. Linearly interpolated reprojection
8. Results

Talk (Anna's part)

Mon Feb 1 16:10:17 MET 1999