GitLab

* ambient occlusion refactored as a decorator
* pre-integrated raycaster and processor to pre-integrate TF

Raycaster is based on the technique described in "Local Ambient Occlusion
in Direct Volume Rendering" by F. Hernell, P. Ljung, A. Ynnermann in IEEE
TVCG 2010.
Implementation does not allow for precomputed LAO volumes but instead
estimates the LAO term per sample.

PipelineRegistry and ProcessorRegistry are now part of campvis-core. Furthermore, moved all calls to PipelineRegistrar<>/SmartProcessorRegistrar<> to a separate cpp file for each module. This way all registrations are at one central location and including headers in external projects does not lead to double registration.

This commit also removes the obsolete columbia and manualsegmentation modules.

refs #386

* Moving to Apache 2.0 license
* Updated AUTHORS.txt

* Introducing SimpleJobProcessor for dispatching non-OpenGL-related jobs

git-svn-id: https://camplinux.in.tum.de/svn/campvis/trunk@445 bb408c1c-ae56-11e1-83d9-df6b3e0c105e

git-svn-id: https://camplinux.in.tum.de/svn/campvis/trunk@338 bb408c1c-ae56-11e1-83d9-df6b3e0c105e

git-svn-id: https://camplinux.in.tum.de/svn/campvis/trunk@335 bb408c1c-ae56-11e1-83d9-df6b3e0c105e

git-svn-id: https://camplinux.in.tum.de/svn/campvis/trunk@326 bb408c1c-ae56-11e1-83d9-df6b3e0c105e

 * added update_legal_header.py script for updating the legal header in all code files of the project
 * ran the file => all TUMVis source files now contain the legal header

git-svn-id: https://camplinux.in.tum.de/svn/campvis/trunk@231 bb408c1c-ae56-11e1-83d9-df6b3e0c105e

 * introduced ReferenceCounted as base class
 * adapted AbstractData, DataHandle and DataContainer

Hence, there is still some work to do to - this version is still not more powerfull than the previous one.

git-svn-id: https://camplinux.in.tum.de/svn/campvis/trunk@201 bb408c1c-ae56-11e1-83d9-df6b3e0c105e

Major revisions to the class layout / data structure:
 * Introduced ImageDataConverter interface (still not really happy with the design)
 * Removed support for Int64 and double images from WeaklyTypedPointer
 * Added ImageDataGL::bind()
 * AbstractProcessor::init() method, gets called by AbstractPipeline::init()
 * added VisualizationProcessor

Updated/New processors:
 * fixed MhdImageReader
 * SliceExtractor stub for very simple slice rendering

Various fixes:
 * DataContaier: managing of DataHandle ownership
 * GenericImageDataLocal::getSubImage()
 * ImageDataRenderTarget bindings
 * GenericProperty
 * linking issues with StringUtils

Hence, all this enables the first usable implementation of a specific pipeline:
The SliceVis pipeline combines MhdImageReader and SliceExtractor for a very simple 2D slice rendering

git-svn-id: https://camplinux.in.tum.de/svn/campvis/trunk@188 bb408c1c-ae56-11e1-83d9-df6b3e0c105e

git-svn-id: https://camplinux.in.tum.de/svn/campvis/trunk@163 bb408c1c-ae56-11e1-83d9-df6b3e0c105e