[bug] small question about on scale different point clouds using s4pcs

[chunibyo-wly]

OpenGR/src/gr/accelerators/pairExtraction/bruteForceFunctor.h

Lines 85 to 112 in 483e4d5

	/*!
	\return Pairs< PointId, PrimitiveId>
	*/
	template <class DummyPrimitive, class Point, int dim, typename Scalar>
	template <class PrimitiveContainer,
	class PointContainer,
	class ProcessingFunctor>
	void
	BruteForceFunctor<DummyPrimitive, Point, dim, Scalar>::process(
	const PrimitiveContainer & M, //!< Input point set \in [0:1]^d
	const PointContainer & Q, //!< Input point set \in [0:1]^d
	Scalar &epsilon, //!< Intersection accuracy, refined
	unsigned int /*minNodeSize*/, //!< Min number of points in nodes
	ProcessingFunctor& functor
	)
	{
	const unsigned int sizeM = M.size();
	const unsigned int sizeQ = Q.size();
	for (unsigned int j = 0; j != sizeQ; ++j){
	functor.beginPrimitiveCollect(j);
	for (unsigned int i = 0; i != sizeM; ++i){
	if( M[i].intersectPoint(Q[j], epsilon ))
	functor.process(i,j);
	}
	functor.endPrimitiveCollect(j);
	}
	}

according to this, I found you set a specific radius when searching in second point cloud.
But when you want to align two point cloud with different scale, I think maybe we can't find a right radius.
So I tried 4pcs and sp4cs with same data.
Here are the results(Tested on hippo model, and resize the second one to 1.2).

s4pcs

/apps/Super4PCS/Super4PCS -i ../assets/hippo1.obj ../assets/hippo2.obj -o 0.7 -d 0.01 -t 1000 -n 200 -r super4pcs_fast.obj -m mat_super4pcs_fast.txt

4pcs

/apps/Super4PCS/Super4PCS -i ../assets/hippo1.obj ../assets/hippo2.obj -o 0.7 -d 0.01 -t 1000 -n 200 -r super4pcs_fast.obj -m mat_super4pcs_fast.txt -x