Difference between revisions of "Mipmapped Volume Rendering"

From Immersive Visualization Lab Wiki
Jump to: navigation, search
(TODO)
(TODO)
 
(9 intermediate revisions by one user not shown)
Line 1: Line 1:
== Long Term Goal ==
 
 
 
==Current Project Status==
 
==Current Project Status==
  
 +
* meshing does not work correctly.  check distance function and coordinate system
 +
* thin volume rendering: pressing texture
 +
* awful image quality..
  
 
==TODO==
 
==TODO==
[[HanProgressReport]]
+
 
# Multi volume support
+
* check meshing error in the cave
# Reimplement Pixel Buffer Object
+
* alpha blending
# Move configuration file parsing from BrickManager class to MeshRend class (done)
+
* image quality problem - preintegration?
# Move duplicated code in QuadInfo and CubeInfo class to BrickInfo super class (done)
+
* scaling
# Reimplement Multi-channel support (done)
+
* 16 bit support
#* modify texture memory allocation to handle single channel
+
* directory selection
#* vvGLSL class
+
* performance check
# Modify downsampler with padding
+
 
#* the current downsampler does not pad empty data when a brick is not of a regular size. Nancy and I implemented in this way because it was easier than padding. However, this causes lots of problems from the reader part, mostly additional checking codelet whether a brick has a regular size, e.g. 128, or smaller than that. Due to the exception, we need couple more variables for boundary bricks of which size is usually smaller than the brick size. By padding, these inefficient and unnecessary data can be removed.
+
 
 +
===Multi volume support===
 +
* For now, it can load multiple volumes but the algorithm used here is very simple: overlaying two rendered images.
 +
* Chih's work hasn't been integrated. I also have an idea about the correct multi volume rendering equation.
 +
 
 +
=== Downsampler ===
 +
* Downsampler has lots of problems and I'd like to completely rewrite the program.  
 +
* OpenGL API will accelerate this process.
 +
* Thin volume support
 +
 
 +
=== Renderer ===
 +
* Removing padding
 +
 
 +
=== Transfer Function ===
 +
* 1D transfer function
 +
 
 +
===Optimization===
 +
* 3D Texture compression?
 +
* Progressive refinement
 +
 
 +
=== Test Set ===
 +
* Collect a set of volume so I can always check after major modification of the code
 +
 
 +
=== Software Release ===
 +
* find an easy workflow minimizing editing
 +
* script
 +
* qmake
 +
 
 +
===Tiled Display===
 +
* works with CGLX but without any optimization
 +
* StarCAVE
 +
* Varrier
 +
 
 +
== Long Term Goal ==
 +
* integrate with deskvox
 +
* integrate with my multi-dimensional transfer function

Latest revision as of 13:59, 27 October 2009

Contents

Current Project Status

  • meshing does not work correctly. check distance function and coordinate system
  • thin volume rendering: pressing texture
  • awful image quality..

TODO

  • check meshing error in the cave
  • alpha blending
  • image quality problem - preintegration?
  • scaling
  • 16 bit support
  • directory selection
  • performance check


Multi volume support

  • For now, it can load multiple volumes but the algorithm used here is very simple: overlaying two rendered images.
  • Chih's work hasn't been integrated. I also have an idea about the correct multi volume rendering equation.

Downsampler

  • Downsampler has lots of problems and I'd like to completely rewrite the program.
  • OpenGL API will accelerate this process.
  • Thin volume support

Renderer

  • Removing padding

Transfer Function

  • 1D transfer function

Optimization

  • 3D Texture compression?
  • Progressive refinement

Test Set

  • Collect a set of volume so I can always check after major modification of the code

Software Release

  • find an easy workflow minimizing editing
  • script
  • qmake

Tiled Display

  • works with CGLX but without any optimization
  • StarCAVE
  • Varrier

Long Term Goal

  • integrate with deskvox
  • integrate with my multi-dimensional transfer function