State of the Art Research

1
State of the Art Research

• JunHyeok Heo
• HyungSeok Kim
• Computer Sciene Dept. of KAIST

2
The State of the Art Culling

• JunHyeok Heo
• Department of Computer Science, KAIST

3
Objective
Culling
Remove perceptually unimportant polygons
Remove invisible polygons
Keep a superset of visible polygons
Conservative Visibility
4
Difficulties
Culling

• There are always some situations which results in
  a negative gain
• The entire scene resides within the view frustum
• Very low depth complexity
• How to maximize gain
• maximize the saved time in rendering minimize
  the number of polygons
• minimize the computational overhead in runtime
• Preprocessing of static parts of scene

5
Culling
6
Classification of Related Work
Culling

• Visibility Preprocessing
• Preprocess enumerating all visible polygons for
  all possible positions where the viewer can
  locate at.
• Application specific solutions
• Hierarchy of scene
• Preprocess making a hierarchy of scene(scene
  graph)
• Have a focus on the runtime part
• Somewhat application independent solutions
• The more runtime overhead, the more sophiscate
  method

7
Visibility Preprocessing
Culling

• Ideal solution
• Enumerating all visible polygons for all possible
  positions where the viewer can locate at.
• Aspect Graph Plantinga 90
• the maximum number of nodes O(n9)
• the construction time O(n9logn)
• impractical because of time and space complexity
• Application specific solutions
• Exploiting the characteristics of models used in
  the specific applicatoins

8
Culling
Aspect Graph
a7
a7
a8
a6
a6
a8
p4
p3
a1
a1
p1
a5
a5
p2
a2
a2
a3
a4
a3
a4
node - cell edge - adjacency of cells
cell - the partitioned viewpoint space
9
Architecture - Indoor
Culling
Visibility Preprocessing

• Characteristics
• Well structured, easy partioning of cells (rooms)
• Within a cell, most parts of scene are invisible
  except some parts visible through portals
  (windows, doors)

10

• How to enumerate visible polygons or cells from a
  cell
• Stocastic ray casting Airley90
• Underestimate

11

• Portal stabing Teller91
• Finding sightline through a sequence of n portals
  in O(n2)
• Enumerate visible cells from a cell
• The number of possible portal sequences rapidly
  increases by the number of portals

12

• Dynamic evaluation Luebke 95
• check visibility through portals on the fly by
  using portals as clipping windows on the image
• Portal Texture Aliaga97
• only draw the cell where the viewer currently
  reside in
• portals are textured with previously rendered
  images

13
Architecture - Outdoor
Culling
Visibility Preprocessing

• Complex urban environments
• Single occluder method Cohen 98
• Partioning with regular grids
• Invisible buildings
• All lines between vertics of the cell and all its
  vertices are blocked by single other building.
• Cant deal the occlusion by several cooperative
  occluders
• Require too small cell to get reasonable culling
  results

14
A cell
occluder
occluder
occludee
occludee
A cell
15

• KAIST VRLab. Heo99 submitted
• Solve the occlusion by several cooperative
  occluders
• Get reasonable culling results with somewhat big
  size of cell
• Control the suitable tradeoff between the
  preprocessing time and the culling result

16
x
y
1
1
x2
y2
top
x1
y1
in
?
0
?
0
1
1
bottom
x2
y2
x1
y1
-1
-1
x-node sub-tree
y-node sub-tree
in
p1
in
ys1
x-node
xs1
p1
top
top
ys2
xs2
y-node
bottom
bottom
ys3
xs3
ternary visibility tree
17
Medicine - Endoscopy
Culling
Visibility Preprocessing

• Cave style models
• Corridor test Yagel96

vacant cell
opaque cell
corridor
18
Hierarchy
Culling

• Back Face Culling
• Remove groups of back faces using a hierarchical
  structure
• View Frusum Culling
• Remove groups of objects out of view frustum
  using a hierarchical structure
• Occlusion Culling
• Remove the occluded parts, too.
• Test the occlusion
• Image Space
• Object Space
• Somewhat application independent...

19
View Frustum Culling
Culling
Hierarchy

• Based on Objects
• Hierarchy of bounding volume
• Previously used to accelerate ray tracing
• Objects as primitives
• Users responsibility for the efficiency
• organize objects
• avoid single objects with many polygons
• (eg.) indoor architeture models
• group objects hierarchically

20

• Tightness of bounding volume
• Bounding box Clark 76
• Deformed bounding box by affine matrix Rubin 80
• Slabs Kay 86
• the more computation require to check against the
  view frustum

21

• Adaptive subdivision of space
• BSP Tree
• Octree
• Garlick et al. 90
• QuadTree
• Terrain - Falby et al. 93 - NPSNet
• Automatic construction of the hierarchy
• Splitting of primitives (objects, polygons)
• associate somewhat big primitives with the
  non-leaf nodes
• associate somewhat small primitives with all
  children nodes
• more than twice rendering
• Polygons as primitives

22
Occlusion Culling
Culling
Hierarchy

• Occlusion Test
• Image space using image pyramid
• the hierarchy on image space
• Object space using critical planes

23
Occlusion Culling on Image space
Culling
Hierarchy

• Hierarchical Z-buffer Greene 93
• Object Space Octree (possibly any hierarchical
  structure)
• Image Space Hierarchical Z-buffer (Z-pyramid)
• Impractical without hardware support for
  constructing Z-pyramid
• difficulty in parallelism most commerical
  graphics system pursue

24
level 0 1 pixel
level n 2n x 2n
25
Culling
Hierarchy

• Hierarchical Polygon Tiling Greene 96
• Object Space Octree of BSP trees
• construct a BSP tree with polygons associated
  with each leaf node of Octree
• strict front-to-back traversing
• Image Space Hierarchical Triage Coverage Mask
• 3 states - partial, vacant, opaque
• Impractical without hardware support for
  constructing Z-pyramid
• difficulty in parallelism

26
Culling
Hierarchy

• Hierarchical Occlusion Maps Zhang97
• Object Space any hierarchical structure
• Image Space Hierarchical Occlusion Maps
• HOM construction using texture mapping
• Practical with machines supporting hardware
  texture mapping
• Conservative early termination or Approximate
  visibility culling
• transparency threshold
• Discrimating occluders between occludees

27
(No Transcript)
28
Occlusion Culling on Image space
Culling
Hierarchy

• Shadow Frusta Hudson 97
• Possibly any hierarchical structure
• Select one large convex occulder
• Shadow casting
• elminate objects within the shadow
• Not consider several cooperative occluders
• Discrimating occluders between occludees

C
A
B
29
Culling
Hierarchy

• Linear Critical Surfaces Coorg 96,97
• Octree possibly any hierarchical structure
• Supporting Planes
• Separating Planes
• Frame coherence
• Not consider several cooperative occluders
• Discrimating occluders between occludees

A
30
Discussion
Culling

• Visibility Preprocessing
• less runtime overhead
• application specific approaches
• Runtime culling algorithms using hierarchy
• more runtime overhead
• application independent
• how to select good occluders

31
Culling

• Selection of occluders
• Distance Zhang97
• Approximated Solid Angle Coorg 96,97
• Is there any way to exploit preprocessing for the
  selection of good occluders

32
Culling

• Dynamic environments
• Types of dynamic objects
• Active objects moving within a bounded volume or
  along a predefined path (eg. merry-go-round)
• gt can be treated same as static objects
• Active objects expected as moving along a
  predicted path
• gt TBV(temporal bounding volume) Sudarsky96
• Active objects moving unpredictably
• gt Update the hierarchy everyframe ???
• gt Using the single bounding volume of
  individual object

a
c
d
b
33
References

• Airley90 Airey, J.M., Increasing Update Rates
  in the Building Walkthrough System with Automatic
  Model-Space Subdivision and Potentially Visible
  Set Calculations, Ph.D. Thesis, UNC Chapel Hill,
  1990.
• Clark76 Clark, J. H., Hierarchical Geometric
  Models for Visible Surface Algorithm,
  Communications of ACM, Vol. 19, No. 10, pp.
  547-554, Oct. 1976.
• Cohen98 Cohen-Or, D., Fibich, G., Halperin, D.,
  and Zadicario, E, Conservative Visibility and
  Strong Occlusion for Viewspace Partitioning of
  Densely Occluded Scenes, EUROGRAPHICS98, 1998.
• Coorg96 Coorg, S., and Teller, S., Temporally
  Coherent Conservative Visibility, ACM Symposium
  on Computational Geometry96, 1996.
• Coorg97 Coorg, S., and Teller, S. Real-Time
  Occlusion Culling for Models with Large
  Occluders, ACM Symposium on Interactive 3D
  Graphics97, 1997.
• Falby93 Falby, J.S., Zyda, M. J., Pratt, D.R.,
  and Mackey, R.L., NPSNET Hierarchical Data
  Structures for Real-Time Three-Dimensional Visual
  Simulation, Computer Graphics, Vol. 17, No. 1,
  pages 65-69, 1993.
• Foley96 Feley, J.D., van Dam, A., Feiner, S.K.,
  and Hughes, J.F., Chapter 15 Visible-Surface
  Determination, Computer Graphics principles and
  practice, pages 649-720, 1996.
• Garlick90 Garlick, B., Baum, D.R., and Winget,
  J.M., Interactive viewing of large geometric
  databases using multiprocessor graphics
  workstations, ACM SIGGRAPH 90 Course Notes
  (Parallel Algorithms and Architectures for 3D
  Image Generation), 1990.
• Georges95 Georges, C., Obscuration Culling on
  Parallel Graphics Architectures, UNC-TR95-017,
  May 1995.

34
References

• Greene93 Greene, N., Kass, M., and Miller, G.,
  Hierarchical Z-Buffer Visibility, ACM SIGGRAPH
  93, pp. 231-238, 1993.
• Greene96 Greene, N. Hierarchical Polygon Tiling
  with Coverage Masks, ACM SIGGRAPH96, pp. 65-74,
  1996.
• Hudson97 Hudson, T., Manocha, D., Cohen, J.,
  Lin, M., Hoff, K., and Zhang, H., Accelerated
  Occlusion Culling using Shadow Frusta, ACM
  Symposium on Computational Geometry, June 1997.
• Kay86 Kay, T.L., Kajiya, J.T., Ray Tracing
  Complex Scenes, ACM SIGGRAPH86, 20(4), pages
  269-278, August 1986.
• Lebke95 Luebke, D., and Georges, C., Portals
  and Mirrors Simple, Fast Evaluation of
  Potentially Visible Sets, ACM SIGGRAPH Special
  Issue on 1995 Symposium on Interactive 3D
  Graphics, pp. 155-162, 1995.
• Rubin80 Rubin, S.M. and Whitted, T, A Three
  Dimensional Representation for Fast Rendering of
  Complex Scenes, ACM SIGGRAPH 80, Vol. 14,No. 3,
  pp. 110-116, July 1980.
• Sudarsky96 Sudarsky, O., and Gotsman, C.,
  Output-Sensitive Visibility Algorithms for
  Dynamic Scenes with Applications to Virtual
  Reality, EUROGRAPHICS96, 1996.
• Teller92 Teller, S. Visibility Computations in
  Densely Occluded Polyhedral Environments, Ph.D.
  Thesis, Technical Report UCB/CSD/92/, Computer
  Science Department, University of California at
  Berkeley, 1992.
• Yagel96 Yagel, Roni and Ray, William,
  Visibility Computation for Efficient Walkthrough
  of Complex Environments, Presence, Vol. 5, No. 1,
  pp. 45-60, MIT Press, 1996.
• Zhang97 Zhang, H., Manocha, D., Hudson, T., and
  Hoff III, K. E., Visibility Culling using
  Hierarchical Occlusion Maps, ACM SIGGRAPH97, pp.
  77-88, 1997.

35
The state of the art LOD

• HyeongSeok Kim
• Department of Computer Science, KAIST

36
Level of Detail

• LOD
• Using multiresolution representation to bound
  processing time in limited hardware spec.
• Basic topics
• Multiresolution model generation
• Select / control the resolution of the model

37
Multiresolution Generation
Part I -
Part I -

• HyungSeok Kim
• Department of Computer Science, KAIST

38
Introduction

• Multiresolution model generation
• Generate multiresolution model from high detail
  representation
• Issues
• High detail representation
• Resolution
• Preserving visual fidelity
• Features on 3D geometry
• Non-geometric features
• Simplification - Detail reducing
• Multiresolution representation

39
Introduction

• High-detail representation
• Scanned data
• Laser scanned data / Medical images /
• Over detailed data
• Re-constructed model with well-formed surfaces
• CAD data
• Architectural model / Models for the
  manufacturing /
• Lot of detail points (With more features)
• Not well formed or organized
• Real-time ready data
• Already simplified for a single component
• Combined scene complexity with a lot of simple
  components

40
Introduction

• Resolution
• High resolution requires more rendering time
• Low resolution requires less rendering time
• Number of polygons
• Other parameters?
• Visual fidelity
• Perceptual effects on the object detail

41
Simplification (1)

• 1st motivation of the simplification
• Excessive detail of the scanned data
• Basic Assumption
• Object representation Polygon mesh with
  material information
• Simplifying polygonal model
• Preserving visual fidelity minimizing error
• Simplifying model with small number of polygons
• Select small number of feature points
• Mesh re-construction

42
Grid Subsampling
1st motivation - Scanned Data

• Selection of representative points on the grid
• Simple - Good for arbitrary models (especially
  for moving objects)
• No adaptation on local variation

43
Error calculation

• Local Error Metrics
• Mesh Decimation
• Mesh Refinement
• Global Error Metrics
• Mesh Optimization
• Geometric Error Bound

44
Decimation Schroeder92
1st motivation - Scanned Data

• Select vertices to simplify using vertex assigned
  error
• Remove vertices
• with errors less
• than the threshold
• Retriangulation
• Fast
• Local criteria

Images excerpt from Schroeder92
45
Triangular subdivision Floriani82,83
1st motivation - Scanned Data

• Insert new point to the initial simple
  representation
• Simple natural hierarchy
• Problem on initial approximation
• Local criteria

46
Point repulsion Turk92
1st motivation - Scanned Data

• Positioning points randomly
• Repositioning points
• using repulsion force
• curvature relative
• strength
• Error bound

Images excerpt from Turk92
47
Mesh optimization Hoppe93
1st motivation - Scanned Data

• Error Minimizing approximated solution

Images excerpt from SIGGRAPH 97 Course Notes
48
Simplification Envelope Cohen96
1st motivation - Scanned Data

• Use of envelope curve
• Adaptive
• Bounding error

Images excerpt from Cohen96
49
Summary on simplification (1)
1st motivation - Scanned Data

• 3D data from the 3D scanner
• Well-formed over-detailed surfaces
• With little of the non geometric data
• Surface simplification preserving
• Overall shapes
• Object topologies
• Result of the simplification
• Set of multiple detail representation for an
  object

50
Summary on simplification (1)

• In real world applications
• Not yet quite popular data sources for the
  real-time rendering
• Good for the show room with small number of
  detailed models
• Difficulties for the CAD-based and real-time
  ready models
• Non-manifold surfaces
• Non-geometric data - Color / Texture

51
Other Issues
Other issues
1st motivation - Scanned Data

• Non-geometric information
• Color, Texture
• Appearance Preserving Simplification Cohen98
• Color simplification Certain96, Hoppe96
• Using texture as a simple representation
  Maciel95Aliaga96
• Realtime texture generation Schaufler96

52
Geometry with color simplificationCertain96
1st motivation - Scanned Data

• Simplifying geometric details and/or color
  details
• Possibility of using textures for the simplified
  geometric regions

Images excerpt from Certain96
53
Appearance Preserving Simplification Cohen98
1st motivation - Scanned Data

• Considering Geometry and Texture/Normals
• Normal/Texture Coordinate Consideration
• Normal/Texture Map

Images excerpt from Cohen98
54
Simplification (2)

• Other motivations of the simplification
• CAD-based models
• Non-manifold surfaces
• With variety of non-geometric data
• For applications like
• Architectural walkthrough / Virtual factory /
  Virtual museum / ...
• Real-time ready models
• Set of enough-simplified models
• Need of inter-object simplification
• For applications like
• Real-time simulations (Flight / Driving) /
  Entertainment / ...

55
Vertex Collapsing Garland97
For other motivations

• Extending edge-collapsing operation to the
  inter-object simplifications
• For arbitrary topology

Images excerpt from Garland97
56
Vertex Clustering Luebke97

• Extending Grid Sub-sampling to the inter-object
  simplification
• Selection of representative vertex
• For arbitrary topology

Images excerpt from Luebke97
57
Portal texture Aliaga97
For other motivations

• Use of a set of textures for the geometry outside
  the portal
• Good for the in-door architectural walkthrough

Images excerpt from Aliaga97
Image
58
Imposter-based Approach
For other motivations

• Build highly simplified imposters
• Use of single Textured imposter instead of actual
  3D model
• Marciel95Schaufler96

Images excerpt from Marciel95
59
Simplicial complexes Popovic97
Images excerpt from Popovic97
60
Summary on simplification (2)

• For the CAD data or real-time ready 3D data
• Non-2 manifold surfaces
• Topologically modifying simplification
• With less possibility of polygon-wise
  simplification
• Use of non-polygonal representations Texture
• Need of the scene-wide consideration
• In real-world applications
• How to represent multiresolution models?
• In usual scene graph
• Other possibilities?

61
Multi-resolution model

• Levels on representation detail
• Representations with different number of polygons
• For use in real-time detail selection
• Discrete level representation
• Continuous level representation
• Selection of appropriate detail in hierarchy
• Smooth transition between two consecutive levels
• Single object vs. Scene
• Multi-resolution model with non-geometric
  information
• The multi-resolution model and the scene graph

62
Discrete level representation

• LOD Nodes
• Easy scene graph integration

Images excerpt from OpenGL OptimizerTM
Programmers Guide
63
Triangular subdivision Floriani92,Lounsbery94

• Begin with base mesh
• successive subdivision
• offset hierarchy (wavelet)
• Topology preserving base mesh Eck95
• Easy scene graph integration

Images excerpt from Lounsbery94
64
Progressive mesh Hoppe96

• Sequence of edge collapsing operation
• Suitable for arbitrary mesh
• Smooth transition
• Easy scene graph integration

Images excerpt from SIGGRAPH 97 Course Notes
65
Need of adaptive simplification

• Adaptive on time varying parameter
• Silhouette
• View frustum

Images excerpt from Hoppe97
66
Vertex tree Xia96, Hoppe97,Luebke97

• Hierarchy on vertices of edge collapsing
  operation
• Adaptive simplification
• Non-trivial scene graph integration for
  scene-wide hierarchies

Image excerpt from Luebke97
Image excerpt from SIGGRAPH 97 Course Notes
67
View Sphere Kim98

• Polygon hierarchy with respect to the directional
  relationship
• Efficient selection of the silhouette parts
• Non-trivial scene graph integration for
  scene-wide hierarchies

Image excerpt from Kim98
68
Hierarchical Imposters
Image excerpt from Marciel95
69
Summary

• Multiresolution hierarchies
• Octree / Quadtree / Pyramid
• Voxel representation Chamberlain96
• Triangular subdivision
• triangular parametrization Floriani92,
  Lounsbery94
• Progressive mesh Hoppe96
• Vertex tree Xia96, Hoppe97, Luebke97
• View Sphere Kim98
• Hierarchical Imposters Marciel95
• Desirable properties
• Continuous level
• Support adaptive selection

70
Summary

• Issues
• Scene hierarchy
• Geometric hierarchy
• Hierarchy with the texture (imposter)
• Scene graph integration
• Simulation - Objects with dynamic behavior
• Object level scene graph
• Kinematic Hierarchy
• Rendering - Multiresolution hierarchy (Static)
• Usually vertex level scene graph
• Scene graph with texture representation
• Spatial (or directional) Hierarchy

71
Detail Selection
Part II -
Part II -

• HyungSeok Kim
• Department of Computer Science, KAIST

72
Introduction
Application-specific processing
Scene processing
Polygon processing
Pixel processing
Scene graph
Display list
Depth buffer
frame buffer
LOD1
criteria
Texture memory
Generation of LODs
LOD2
LODk
method
LODn
The control of LODs
subject
73
Control subjects

• Subject domain
• Object level
• Scene level
• Subject type
• geometry
• texture
• geometry texture
• shading
• geometry shading
• geometry texture shading

74
Control criterion

• Purpose
• Giving Fast Regular frame rates
• Keeping high visual fidelity
• Criterion
• Object Level
• Silhouette
• Vertex deviation in the image plane
• Scene Level

75

• Scene Level Criterion
• Object size
• Bounding box, bounding sphere
• Object distance
• Distance from the view
• Size on the image plane
• Gaze direction
• Using visual acuity
• Apparent motion
• Object brightness
• Semantics

76
Control methodsFunkhouser93

• Static method
• Select one level from the pre-defined levels,
  based on a simple criteria.
• good for the discrete-level representation.
• most popular - e.g., computer games
• Pros
• easy to implement, fast
• Cons
• impossible to regularize the frame rate.

77

• Feedback method
• The detail level is determined by the previous
  frames.
• Pros
• simple to implement
• Cons
• perform poorly when the frame coherence is weak.
• overshoot, oscillate, diverge
• often used in flight simulators.

78

• Optimization
• Use the estimated rendering time
• Assign detail level within a bound of the
  estimated rendering time
• Usually modeled with
• Benefit (O,L,R) Visual benefit of the object O
  when rendered in detail level L with rendering
  algorithm R
• Cost (O,L,R) Rendering time
• Issues in optimization
• Computing Cost and Benefit

79
Estimation

• Funkhouser 93
• Cost(O,L,R)
• an estimate of the time required to render
  (O,L,R)
• ??? ?? ???? ??? ??
• per primitive coordinate transform., clipping ?
• per pixel rasterization, z-buffering, texture
  mapping ?
• Benefit(O,L,R)
• contribution to model perception
• heuristics
• size, accuracy of rendering algorithm, semantics,
  focus, motion blur, hysteresis

80
Other selection methods

• Hierarchical detail level assignment
• Scene level hierarchy
• Pre-render
• Calculate actual benefits and costs.
• Using frame coherence
• Not to calculate (assign) detail level for every
  frames

81
Summary

• Issues in the control of level-of-details
• Selection criterion
• Selection target
• Selection method
• Issues in the real world applications
• Time required for the selection
• LOD gain in time
• Representations
• Geometric scene graph
• Other advanced representations
• Must be considered in the context of the entire
  pipeline.

82
References

• Aliaga96 D. G. Aliaga, Visualization of Complex
  Models Using Dynamic Texture-based
  Simplification, In proceedings of Visualization
  '96, pp. 101-106, 1996
• Aliaga97 D. G. Aliaga, A.A. Lastra,
  Architectural Walkthroughs Using Portal Textures,
  IEEE Visualization '97, pp. 355-362, Oct. , 1997
• Certain96 A. Certain, J. Popovic, T. DeRose, T.
  Duchamp, D. Salesin, W. Stuetzle, Interactive
  Multiresolution Surface Viewing, In proceedings
  of ACM SGGRAPH '96, pp. 91-98, 1996
• Cohen96 J. Cohen, A. Varshney, D. Manocha, G.
  Turk, H. Weber, P. Agarwal, F. Brooks, W. Wright,
  Simplification Envelops, In proceedings of ACM
  SIGGRAPH '96, pp. 109-118, 1996
• Cohen98 J. Cohen, M. Olano, D. Manocha,
  Appearance-Preserving Simplification, In
  proceedings of ACM SIGGRAPH '98, pp. 115-122,
  1998
• Eck95 Matthias Eck, Tony DeRose, Tom Duchamp,
  Hugues Hoppe, Michael Lounsbery, Werner Stuetzle,
  Multiresoluiton Analysis of Arbitrary Meshes, In
  proceedings of ACM SIGGRAPH '95, pp. 173-182
• Floriani83 Leila De Floriani, Bianca
  Falcidieno, and Caterina Pienovi. Delaunay-based
  method for surface approximation. Proceedings of
  Eurographics'83, pp.333-350, 1983.
• Funkhouser93 Thomas A. Funkhouser, Carlo H.
  Sequin, Adaptive Display Algorithm for
  Interactive Frame Rates During Visualization of
  Complex Virtual Environments, In proceedings of
  ACM SIGGRAPH '93, 1993
• Garland97 M. Garland, P. S. Heckbert, Surface
  Simplification Using Quadric Error Metrics, In
  proceedings of ACM SIGGRAPH '97, pp. 209-216,
  1997

83
References

• Hoppe93 H. Hoppe, T. DeRose, T. Duchamp, J.
  McDonald, W. Stuetzle, Mesh optimization, In
  proceedings of ACM SIGGRAPH '93, pp. 19-26, 1993
• Hoppe96 Hugues Hoppe, Progressive Meshes, In
  proceedings of ACM SIGGRAPH '96, pp. 99-108
• Hoppe97 H. Hoppe, View-dependent Refinement of
  Progressive Meshes, In proceedings of ACM
  SIGGRAPH '97, pp. 189-198, 1997
• Kim98 HyungSeok Kim, SoonKi Jung, KwangYun
  Wohn, A multiresolution control method using view
  directional feature, In proceedings of ACM VRST
  '98, 1998
• Luebke97 D. Luebke, C. Erikson, View-dependent
  Simplification of Arbitrary Polygonal
  Environments, In proceedings of ACM SIGGRAPH '97,
  pp. 199-208, 1997
• Maciel95 P.W.C. Maciel, P. Shirley, Visual
  Navigation of Large Environments Using Textured
  Clusters, In proceedings of Symposium on
  Interactive 3D Graphics, pp. 95-102, 1995
• Popovic97 J. Popovic, H. Hoppe, Progressive
  Simplicial Complexes, In proceedings of ACM
  SIGGRAPH '97, 1997
• Schaufler96 G. Schaufler, Exploiting
  Frame-to-Frame Coherence in a Virtual Reality
  System, In proceedings of VRAIS '96, pp. 95-102,
  1996
• Schroeder92 W.J. Schroeder, J.A. Zarge and W.E.
  Lorensen, Devimation of Triangle Meshes, SIGGRAPH
  92, pp. 65-70, 1992
• Turk92 G. Turk, Re-tiling polygonal surface, In
  proceedings of ACM SIGGRAPH '92, pp.289-294, 1992
• Xia96 Julie C. Xia, Amitabh Varshney, Dynamic
  View-Dependent Simplification for Polygonal
  Models, In proceeding of Visualization '96, pp.
  327-334, 1996