Skip to content

Conversation

@kaspermarstal
Copy link
Member

This PR adds an intensity penalty metric and a shape penalty metric inspired by Active Appearance Models. The metrics are disabled by default.

Kasper Marstal and Stefan Klein "Active registration models", Proc. SPIE 10133, Medical Imaging 2017: Image Processing, 101330Y (24 February 2017); https://doi.org/10.1117/12.2254356).

Kasper Marstal and Stefan Klein "Active registration models",
Proc. SPIE 10133, Medical Imaging 2017: Image Processing, 101330Y
(24 February 2017); https://doi.org/10.1117/12.2254356)
@kaspermarstal
Copy link
Member Author

Mhm, I disabled the metrics by default because of the external dependency on boost. Seems like it still tries to find boost even though the metrics are disabled. Will fix this.

}

template <class F>
typename boost::result_of<F()>::type callstatismoImpl(F f) const {
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Maybe decltype(F()), instead of boost::result_of?

float noiseVariance,
double modelVarianceRetained
) {
statismo::StatisticalModel<Representer>* model_statismo = callstatismoImpl(boost::bind(&ImplType::BuildNewModel, this->m_impl, SampleDataStructureList, surrogateTypes, conditioningInfo, noiseVariance, modelVarianceRetained));
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Instead of boost::bind, maybe use std::bind...?

typedef MeshType DatasetType;

// An unordered map is used to cache pointid for corresonding points
typedef boost::unordered_map<PointType, unsigned> PointCacheType;
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Instead of boost::unordered_map, maybe use std::unordered_map...?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants