Next: 4.3 Data preparation
Up: 4 Theory
Previous: 4.1 Basic Idea
  Contents
Volume rendering is one of the most popular algorithms for volume visualization.
It is an image-based approach, i.e. a view and an image plane are set for which
the projection of the volume is done pixel-by-pixel.
Volume rendering consists of two main parts. First, some pre-processing computations are done,
to classify and shade each voxel of the volume : an opacity value is obtained
for the classification and a colour value is obtained for the shading.
Then, it is followed by the actual rendering part, i.e. the projection from three
dimensional data to a two dimensional image.
The volume rendering pipeline is usually consists of :
- data preparation (pre-filtering, resampling, maybe segmentation)
- gradient computation
- classification by assigning opacity
- shading by assigning colour
- compositing and projection by ray casting
Elodie Fourquet
2005-01-18