Clone wiki

STA-ImageToolbox / Home

(added Oct 2014) We made a newer version of the sta-toolbox publicly available. There is now experimental "native matlab support". This means, it should be possible to compile most of the code without installing the gsl or fftw (particularly useful for windows users).

Welcome

The current version of the STA-ImageToolbox provides matlab & c++ code for densely computing local
3D rotation invariant image descriptors based on our paper:

Henrik Skibbe, Marco Reisert, Thorsten Schmidt, Thomas Brox, Olaf Ronneberger and Hans Burkhardt,
"Fast Rotation Invariant 3D Feature Computation utilizing Efficient Local Neighborhood Operators"
in IEEE Transactions on Pattern Analysis and Machine Intelligence, accepted. (Preprint available here)

If you use our functions or partially make use of our code then please cite this paper.

this is my image

Feel free to check out the code or download a ZIP file containing the most current version.

licence

GPL 3.0

Requirements

installation instructions can be found in the README.txt file in the root directory
of the repository.

successfully tested with
windows 7 64bitoctave 3.4.3
ubuntu 11.10 64bitmatlab 2010a
ubuntu 12.04 64bitmatlab 2011

We currently offer the following functions (matlab & octave)

function name
sta_localfeat_SLD_cfastspherical Laguerre Descriptor SLDpreferred implementation
sta_localfeat_SLD_cspherical Laguerre Descriptor SLD
sta_localfeat_SGD_cfastspherical Gabor Descriptor SGDpreferred implementation

matlab only:

function name
sta_localfeat_SLD_mspherical Laguerre Descriptor SLD

Bibliography

Henrik Skibbe, Marco Reisert, Thorsten Schmidt, Thomas Brox, Olaf Ronneberger and Hans Burkhardt,
"Fast Rotation Invariant 3D Feature Computation utilizing Efficient Local Neighborhood Operators"
in IEEE Transactions on Pattern Analysis and Machine Intelligence, accepted

Updated