Range image segmentation and fitting by residual consensus

The authors randomly sample appropriate range image points and solve equations determined by these points for the parameters of selected primitive type. From K samples they measure residual consensus to choose one set of sample points that determines an equation having the best fit for the largest homogeneous surface patch in the current processing region. The residual consensus is measured by a compressed histogram method that works at various noise levels. The estimated surface patch is extracted out of the processing region to avoid further computation. A genetic algorithm is used to accelerate the search speed.<<ETX>>