A benchmark for 3D mesh segmentation

Chen, Xiaobai; Golovinskiy, Aleksey; Funkhouser, Thomas

doi:10.1145/1576246.1531379

Cited by 291 publications

(586 citation statements)

References 32 publications

Supporting

Mentioning

570

Contrasting

Unclassified

Order By: Relevance

“…11. We get good performances in the man, hand, dolphin and many other models provided by X. Chen et al [8]. They are decomposed to meaningful components without over-segmentation and remain good boundaries between each component.…”

Section: A Some Resultsmentioning

confidence: 81%

Mesh Decomposition Using Skeleton and Fuzzy Clustering

Wu¹,

Xiao²

2013

IJMLC

View full text Add to dashboard Cite

Abstract-As 3D models are used more and more in various applications, model decomposition acts as an important step for model understanding and can be used as a pre-processing step for different applications. This paper presents a mesh decomposition method based on skeleton. In recent years, a lot of work has been done to decompose polygon meshes to sub-meshes. No standard for the result, but the aim is to get reasonable sub-meshes and keep the boundaries between them fine. In this paper, we propose a mesh decomposition algorithm using skeleton and fuzzy clustering. Our method uses the skeleton of the mesh as the base to automatically get the meaningful components without over-segmentation and keeps a fine boundary using fuzzy clustering.

show abstract

Section: A Some Resultsmentioning

confidence: 81%

Mesh Decomposition Using Skeleton and Fuzzy Clustering

Wu¹,

Xiao²

2013

IJMLC

View full text Add to dashboard Cite

show abstract

“…In this paper we chose to adopt the dataset and metrics proposed by Chen et al (2009). This choice was driven by three factors: the richness of the dataset in terms of number of meshes and categories, the presence of a ground-truth segmentation operated by a number of different individuals and, finally, the availability of a large set of results obtained by running most modern 3D mesh segmentation techniques with this benchmark.…”

Section: Experimental Validationmentioning

confidence: 99%

“…In order to assess the results obtained by the proposed technique we first performed a set of quantitative experiments over the dataset presented by Chen et al (2009).…”

Section: Quantitative Evaluationmentioning

confidence: 99%

“…3), that forces all local refinements to be in the same direction, and the Local Consistency Error (LCE in Fig. 3), that allows for different refinement directions in different parts of the same object (refer to (Chen et al, 2009) for more details about these metrics). The compared method were Randomized and Normalized Cuts (Golovinskiy and Funkhouser, 2008), Shape Diameter Functions (Shapira et al, 2008), Core Extraction (Katz et al, 2005), Random Walks (Lai et al, 2008), Fitting Primitives (Attene et al, 2006a) and K-Means (Shlafman et al, 2002).…”

Section: Quantitative Evaluationmentioning

confidence: 99%

“…For each approach we used the optimal parameter set suggested by Chen et al (2009). In addition, a set of totally humansupervised segmentation is available in the benchmark.…”

Section: Quantitative Evaluationmentioning

confidence: 99%

See 2 more Smart Citations

A graph-based technique for semi-supervised segmentation of 3D surfaces

Bergamasco

Albarelli

Torsello

2012

Pattern Recognition Letters

View full text Add to dashboard Cite

a b s t r a c tA wide range of cheap and simple to use 3D scanning devices has recently been introduced in the market. These tools are no longer addressed to research labs and highly skilled professionals, but rather, they are mostly designed to allow inexperienced users to acquire surfaces and whole objects easily and independently. In this scenario, the demand for automatic or semi-automatic algorithms for 3D data processing is increasing. In this paper we address the task of segmenting the acquired surfaces into perceptually relevant parts. Such a problem is well known to be ill-defined both for 2D images and 3D objects, as even with a perfect understanding of the scene, many different and incompatible semantic or syntactic segmentations can exist together. For this reason recent years have seen a great research effort into semi-supervised approaches, that can make use of small bits of information provided by the user to attain better accuracy. We propose a semi-supervised procedure that exploits an initial set of seeds selected by the user. In our framework segmentation happens by propagating part labels over a weighted graph representation of the surface directly derived from its triangulated mesh. The assignment of each element is driven by a greedy approach that accounts for the curvature between adjacent triangles. The proposed technique does not require to perform edge detection or to fit parametrized surfaces and its implementation is very straightforward. Still, despite its simplicity, tests made on a standard database of scanned 3D objects show its effectiveness even with moderate user supervision.

show abstract

User‐guided volumetric approximation using swept sphere volumes for physically based animation

Bae

Kim

2012

Computer Animation & Virtual

View full text Add to dashboard Cite

We present an efficient, user-guided volumetric approximation algorithm, specifically designed for physically based animation. Our method combines automatic and interactive segmentation methods to give users an intuitive and easy way to approximate 3D meshes. Our approach first constructs the simplified medial axis transform of the input mesh object, and segments the medial axis into parts in terms of swept sphere volumes using a region growing method. Then, we decompose the object surface into regions based on the mapping between the segmented medial axis and the object surface. Each segmented region is approximated with a swept sphere volume. These decomposed surface regions can be interactively refined further by splitting and/or merging using a sketch-based input. Experimental results show that our approach produces good volumetric approximation results for different types of object shapes. Moreover, rigid-body dynamics simulation based on our volumetric approximation provides a visually pleasing result.

show abstract

A benchmark for 3D mesh segmentation

Cited by 291 publications

References 32 publications

Mesh Decomposition Using Skeleton and Fuzzy Clustering

Mesh Decomposition Using Skeleton and Fuzzy Clustering

A graph-based technique for semi-supervised segmentation of 3D surfaces

User‐guided volumetric approximation using swept sphere volumes for physically based animation

Contact Info

Product

Resources

About