Segmentation of point clouds using smoothness constraints

By Tahir Rabbani, Frank van den Heuvel, George Vosselman. 2006

445 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/79c9aa9b15fd18bd510f99ec8b6fc63802dcc033

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

A Global Hypotheses Verification Method for 3D Object Recognition (2012, methodology): “To counterattack the effect of clutter, we devised an approach, inspired by surface-based segmentation [13], aimed at penalizing a hypothesis that locally explains some part of the scene but not nearby points belonging to the same smooth surface patch.”
Automatic Creation of Semantically Rich 3D Building Models from Laser Scanner Data (2011, methodology): “The region-growing algorithm is similar to (Rabbani et al., 2006).”
SegMatch: Segment based place recognition in 3D point clouds (2016, methodology): “this module can be replaced by a different algorithm. Fig. 7 shows an example of a correct loop detection by matching segments obtained from segmenting the point cloud based on smoothness constraints [30]. Although these types of segments may appear to be less meaningful for humans, they provide dis”
AN ADAPTIVE APPROACH FOR SEGMENTATION OF 3D LASER POINT CLOUD (2012, methodology): “Rabbani et al. (2006) utilized the surface roughness to group points belonging to the same surface.”
Review: deep learning on 3D point clouds (2020, unknown): “Traditionally, segmentation is done using edges [139] or surface properties such as normals, curvature and orientation [139,140].”
Segmentation of 3D lidar data in non-flat urban environments using a local convexity criterion (2009, methodology): “Attributes used inc lude surface direction [13], curvature [14], [15], edges[16], l ocal smoothness constraints [ 17 ] and more.”
Min-cut based segmentation of point clouds (2009, background): “What work exists mostly focuses on the extraction of geometric primitives or parts( [13, 18]) rather than entire objects.”
Persistent Point Feature Histograms for 3D Point Clouds (2008, background): “Estimated surface curvatures and normals for a point [1] are two of the most widely used point features, and play an important role in applications such as registration [2], or segmentation [ 3 ].”
Shape-based recognition of 3D point clouds in urban environments (2009, background): “Algorithms have been proposed for modeling specific object types (e.g., buildings [2, 9] and trees [28, 30]), for extracting geometric primitives (e.g., [ 22 , 25]), and for identifying objects in cluttered scenes [3, 15, 10].”