Realtime visualization of large detailed volumes on GPU

1
Real-time visualization of large detailed volumes
on GPU
Interactive GigaVoxels

• Cyril Crassin, Fabrice Neyret, Sylvain Lefebvre
• INRIA Rhône-Alpes / Grenoble Universities

2
Volumetric representations for special effects

• Volumetric special effects KH05, KAP02
• Impressive natural scenes
• Accurate director vision modeling
• Realistic and controllable model
• Voxel engines
• Realistic global illumination models
• Effects compositing
• Particles, fluids, etc.
• Unified rendering
• Costly rendering solutions
• Computations time
• Memory occupation

Source Digital Domain
3
What we do

• Voxels ray-casting for real-time scenes in Video
  Games
• Full scenes representation
• Multi-Scale Large and detailed
• Hyper-texture usage
• Details enhancement on surface-based scenes
• Procedural amplification
• Advantages
• Compact representation for very small details
• Memory and rendering efficiency
• Easy filtering and LOD
• 3D Mip-Mapping
• A lot of tricky effects becomes easy
• Depth Of Field, volumetric light effects

4
Pictures and videos
5
Data structure

• Hybrid data structure
• N3-tree Generalized Octree
• Allow multi-scale subdivision depending on data
  densities
• Limit tree depth for rendering performances
• Acceleration structure
• Empty space skipping
• Dense, empty and detailed zones encoding
• Small voxel grids in leafs
• Rendering performances
• Allow hardware filtering (Tri/Quadri-linear)
• Mip-Mapping
• Memory efficiency
• Improve storage cost/structure cost efficiency
• Local density hypothesis in complex zones
• Data generation and transfer efficiency
• Regular blocs generation/loading
• Bloc transfers to the GPU

6
Rendering

• Screen space volume ray-casting
• Direct N3-tree traversal on GPU
• Close to GPU Kd-tree traversal for triangle
  ray-tracing EVG04,FS05,HSHH07
• Kd-Restart is the most efficient on current
  architectures (G80/G92/GT200)
• Regular bricks rendering in leafs
• Volumetric ray-casting KW03, Sch05
• Hardware filtering
• Mip-Mapping
• Adaptive sampling
• Distance dependant
• Advantages
• Good scaling on huge volumes
• Low dependency on data volume, low overdraw, few
  geometrical manipulation
• Early rays termination
• Lighting
• Classical Blinn-Phong per ray sample
• On the fly computed or stored gradient data
• Shadow maps

7
Out-of-core data streaming (1)

• Produce / Store on GPU only data needed for the
  current point of view
• Tree update and bricks loading in real time
  during exploration
• GPU Cache mechanism
• Pools of chunks
• Implemented in texture memory
• Sub-Texture update operations
• Manually managed LRU mechanism
• Time stamps updated during rendering
• Via visibility mechanism
• Tree storage Nodes-Pool
• Classical pointer-based structure
• Small size 3D texture
• Bricks storage Bricks-Pool
• Very large 3D texture
• Usually the whole remaining GPU memory

8
Out-of-core data streaming (2)

• Visibility based loading and LOD
• Loaded tree branches subdivision/merging
• Depending on brick distance to the view
• A voxel project to one pixel constraint
• Nodes visibility detection
• Per ray needed nodes information provided by
  rendering
• Screen space Nodes ID buffers
• Fast stream compaction operation
• Reduced Nodes ID buffer read back to CPU
• Progressive loading
• Upper nodes used while waiting for new data
• Bricks kept in parents nodes
• Min of 3 levels kept for mip-mapping
  implementation

9
Results

• Scenes
• Lizard
• 20483 SP FLOAT, 32GB
• Visible Man
• 20483 RGBA8, 32GB
• Bones field
• 81923 ALPHA8, 512GB (1GB on disk)
• Sierpinski sponge
• 8.4M3 ALPHA8
• Clouds
• 5123 SP FLOAT
• Hyper-texture usage

10
Visible Man

• 20483 RGBA8, 32GB, 20FPS_at_5122

11
Lizard

• 20483 SP FLOAT, 32GB, 15FPS_at_5122

12
Sierpinski sponge

• 8.4M3 ALPHA8 procedural, 80FPS_at_5123

13
Bones Field

• 81923 ALPHA8, 512GB (1GB on disk)

14
Clouds

• 5123 SP FLOAT

15
G80 Reverse Engineering

• Results
• Dzdz
• See online
• http//www.icare3d.org/GPU/CN08