$next$ $up$ $previous$
: 参考文献 : Unification of Appearance Models : Results

Conclusions and Contributions

We have illustrated not only that group-wise registration based on appearance model is possible, but also that it surpasses registration that is based purely on a reference image.

The current algorithms are being interpreted rather than compiled and no multi-scale approach is yet in use. The extension of the algorithms to 2-D and 3-D would require a long time to run, but remains practical. Compiled implementations should be available within months as well as heuristic optimisations that will make run-time more competitive with that of pair-wise approaches. Furthermore, the results have been shown to be better in a global sense and are not dependent on just one individual image.

Contributions of the work can be subdivided into three aspects:

It provides a benchmark environment and results for many methods, including several new ones.
Unprecedented model-based group-wise registration is introduced.
Automatic construction of increasingly better appearance model becomes practicable. Correspondences are obtained using techniques borrowed from image registration.

Acknowledgements. The project is funded by <<body>> and forms a part of the <<research>>. Contribution to this work and important discussions involved <<anon>> and <<anon>>. A significant proportion of the application's code were worked on by <<anon>>.

$next$ $up$ $previous$
: 参考文献 : Unification of Appearance Models : Results

Roy Schestowitz 平成17年6月5日