Distance field guided $$L_1$$ L 1 -median skeleton extraction

Song, Chengfang; Pang, Zhiqiang; Jing, Xiao‐Yuan; Xiao, Chunxia

doi:10.1007/s00371-016-1331-z

Cited by 24 publications

(25 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We also compared our approach with the three chosen methods for models with incomplete and significant missing data. As shown in Figure 24, compared to the L 1 ‐medial method [HWCO*13] and the distance field guided L 1 ‐median method [SPJX18], we found the skeletons by our method had more details in the head of the dog model (i.e. two ears are represented by two skeletal branches).…”

Section: Resultsmentioning

confidence: 81%

“…We also compared our method with the L 1 ‐medial method [HWCO*13] for two point cloud models in Figure 21, listed in Table 3. In the original works of [SPJX18, HWCO*13], the authors reported that the L 1 ‐medial method [HWCO*13] is faster than the ROSA method [TZCO09] and the distance field guided L 1 ‐median method [SPJX18], while our method is significantly faster than the L 1 ‐medial method [HWCO*13] (refer to Table 3). For the complexity of the algorithm, although we add the normal optimization and the tensor‐based spectral clustering, the computational efficiency of our skeleton extraction method is higher than the L 1 ‐medial method because our method does not need iterations to obtain the approximated centroid points.…”

Section: Discussionmentioning

confidence: 91%

“…Comparisons among the three methods ( L 1 ‐medial [HWCO*13], Distance field guided L 1 ‐median [SPJX18], our method) for a dog model. The number of points is 18 114, and the sampling density is 102.1069.…”

Section: Resultsmentioning

confidence: 99%

“…The minimal distances of all the voxels together constitute the distance field. However, it is computationally expensive to compute a high‐resolution distance field, which limits its application for point cloud models [SPJX18].…”

Section: Initial Skeletal Points Constructionmentioning

confidence: 99%

“…In these cases, the previous algorithms could not be used. Therefore, researchers have proposed several skeleton extraction algorithms from raw scan data, including the generalized rotational symmetry axis (ROSA)‐based method [TZCO09], the L 1 ‐medial skeleton extraction algorithm based on random sampling [HWCO*13] and the combination of the distance field and L 1 ‐median skeleton extraction from raw scanned point cloud data [SPJX18]. How to obtain a centred skeleton with topology preservation from incomplete or noisy point cloud data remains a challenge.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Curve Skeleton Extraction From 3D Point Clouds Through Hybrid Feature Point Shifting and Clustering

Zhong

Jin

et al. 2020

Computer Graphics Forum

View full text Add to dashboard Cite

Curve skeleton is an important shape descriptor with many potential applications in computer graphics, visualization and machine intelligence. We present a curve skeleton expression based on the set of the cross‐section centroids from a point cloud model and propose a corresponding extraction approach. We first provide the substitution of a distance field for a 3D point cloud model, and then combine it with curvatures to capture hybrid feature points. By introducing relevant facets and points, we shift these hybrid feature points along the skeleton‐guided normal directions to approach local centroids, simplify them through a tensor‐based spectral clustering and finally connect them to form a primary connected curve skeleton. Furthermore, we refine the primary skeleton through pruning, trimming and smoothing. We compared our results with several state‐of‐the‐art algorithms including the rotational symmetry axis (ROSA) and L1‐medial methods for incomplete point cloud data to evaluate the effectiveness and accuracy of our method.

show abstract

Section: Resultsmentioning

confidence: 81%