A Hierarchical Segmentation of Articulated Bodies

Goes, Fernando de; Goldenstein, Siome; Velho, Luiz

doi:10.1111/j.1467-8659.2008.01274.x

Cited by 113 publications

(103 citation statements)

References 23 publications

Supporting

Mentioning

103

Contrasting

Order By: Relevance

“…They use techniques such as K-means [30], graph cuts [12], hierarchical clustering [7,8,11], random walks [16], core extraction [13], tubular primitive extraction [21], spectral clustering [19], and critical point analysis [18]. While some research tries to use segmentation criteria that are consistent between meshes in a class (such as shape diameter function [29]) or between articulated versions of a model (such as diffusion distance [5]), it is difficult in general to run segmentation methods independently on a set of meshes and obtain results with corresponding segments ( Figure 1a). We expand such segmentation techniques to simultaneously segment a set of meshes.…”

Section: Previous Workmentioning

confidence: 99%

Consistent segmentation of 3D models

Golovinskiy

Funkhouser

2009

Computers & Graphics

176

149

View full text Add to dashboard Cite

This paper proposes a method to segment a set of models consistently. The method simultaneously segments models and creates correspondences between segments. First, a graph is constructed whose nodes represent the faces of every mesh, and whose edges connect adjacent faces within a mesh and corresponding faces in different meshes. Second, a consistent segmentation is created by clustering this graph, allowing for outlier segments that are not present in every mesh. The method is demonstrated for several classes of objects and used for two applications: symmetric segmentation and segmentation transfer.

show abstract

Section: Previous Workmentioning

confidence: 99%

Consistent segmentation of 3D models

Golovinskiy

Funkhouser

2009

Computers & Graphics

176

149

View full text Add to dashboard Cite

show abstract

“…Finally another more recent work [18] needs to be mentioned. It describes the use of diffusion distances for pose invariant hierarchical shape segmentation.…”

Section: Related Workmentioning

confidence: 99%

“…Furthermore, these methods need to compute several eigenfunctions until a cut-off is reached and it remains unclear how accurately these functions can be computed with the linear discretizations of the Laplace operator used. Nevertheless, [18] shows some impressive hierarchical segmentations. Even though they detect the features on coarser resolutions, they do not necessarily align to the concavities and also segment the shape at locations without specific features on higher resolution levels.…”

Section: Related Workmentioning

confidence: 99%

Hierarchical Shape Segmentation and Registration via Topological Features of Laplace-Beltrami Eigenfunctions

Reuter

2009

Int J Comput Vis

153

View full text Add to dashboard Cite

This work introduces a method to hierarchically segment articulated shapes into meaningful parts and to register these parts across populations of nearisometric shapes (e.g. head, arms, legs and fingers of humans in different body postures). The method exploits the isometry invariance of eigenfunctions of the Laplace-Beltrami operator and uses topological features (level sets at important saddles) for the segmentation. Concepts from persistent homology are employed for a hierarchical representation, for the elimination of topological noise and for the comparison of eigenfunctions. The obtained parts can be registered via their spectral embedding across a population of near isometric shapes. This work also presents the highly accurate computation of eigenfunctions and eigenvalues with cubic finite elements on triangle meshes and discusses the construction of persistence diagrams from the MorseSmale complex as well as the relation to size functions.

show abstract

“…In recent years, we have seen an explosive growth of the available 3D model data across a variety of fields, such as reverse engineering and 3D medical imaging, with the development of the acquisition techniques [1,23,20,16,15,8,7,5,10,18]. We are therefore faced with an ever-increasing demand for approaches towards automatic model processing, understanding and analysis.…”

Section: Introductionmentioning

confidence: 99%

“…We define a segmentation, which can robustly provide a solution to the above four challenges above, as the perceptually consistent mesh segmentation (PCMS). Over the past several years, the integrated characteristics of PCMS has become more important for advanced mesh processing and understanding as it provides more insights into mesh models [8,16,23,25]. PCMS facilitates the interpretation of 3D surface meshes in terms of either a pure geometric sense, semantic information or both through the representation of an intrinsically hidden geometric structure of the meshes.…”

Section: Introductionmentioning

confidence: 99%