RT Journal Article
T1 Hierarchical segmentation of range images inside the combinatorial pyramid
A1 Marfil-Robles, Rebeca
A1 Antúnez, Esther
A1 Bandera-Rubio, Antonio Jesús
K1 Imágenes
K1 Neurociencia computacional
AB RGB-D cameras are not only able to provide color (Red-Green-Blue -RGB-) informa-tion from the scene but also a relatively accurate cloud of 3D points. Using informationcoming from this organized cloud, it is possible to define around each image pixel a smallplanar patch and obtain its normal vector. Within the framework of the combinatorialpyramid, this paper describes a method to abstract from these normals to paramet-ric surface models. The method works at two consecutive stages. Firstly, normalsare hierarchically grouped to divide up the image into superpixels. These superpixelscapture small patches on the scene that belong to the same surface. Then, they aremerged to segment the scene into simple geometric models. Curvature informationand model information are used to divide up the image into planes, cylinders and/orspheres. This paper shows how, in the higher levels of abstraction of the combinatorialpyramid, scenes can be described using these geometric items and their topologicalrelationships.
PB Elsevier
YR 2015
FD 2015
LK https://hdl.handle.net/10630/34513
UL https://hdl.handle.net/10630/34513
LA eng
NO Marfil, R., Antúnez, E., Bandera, A. Hierarchical segmentation of range images inside the combinatorial pyramid. Neurocomputing, 2015, 161, pp. 81–88
DS RIUMA. Repositorio Institucional de la Universidad de Málaga
RD 20 ene 2026