Proceedings 15th International Conference on Pattern Recognition. ICPR-2000
DOI: 10.1109/icpr.2000.905491
|View full text |Cite
|
Sign up to set email alerts
|

Segmentation and surface characterization of arbitrary 3D meshes for object reconstruction and recognition

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
6
0

Publication Types

Select...
6

Relationship

1
5

Authors

Journals

citations
Cited by 11 publications
(6 citation statements)
references
References 4 publications
0
6
0
Order By: Relevance
“…With the use of an appropriate criterion to restrict the matching to certain directions, the trivial solutions of the unconstrained matching are avoided. Such a criterion is suggested in [14] where arbitrary meshes are segmented into crude compact facets and the algorithm marks as candidate directions for matching the average normal vectors of the most irregular facets. This algorithm is well suited for 3D scanned objects and performs particularly well on types of fragments encountered in archaeological applications.…”
Section: Constrained Matching Resultsmentioning
confidence: 99%
See 1 more Smart Citation
“…With the use of an appropriate criterion to restrict the matching to certain directions, the trivial solutions of the unconstrained matching are avoided. Such a criterion is suggested in [14] where arbitrary meshes are segmented into crude compact facets and the algorithm marks as candidate directions for matching the average normal vectors of the most irregular facets. This algorithm is well suited for 3D scanned objects and performs particularly well on types of fragments encountered in archaeological applications.…”
Section: Constrained Matching Resultsmentioning
confidence: 99%
“…Overall, we were satisfied with the results and we were encouraged to apply our complementary matching method to problems where more than one target object must be reconstructed from multiple parts. Such a method is proposed in [13] where the direction-constrained matching and the direction selection criterion of [14] are used for the matching error evaluation between each pair of fragments. Additionally, the maximization of surface overlap and the material axis are used as biasing constraints.…”
Section: Constrained Matching Resultsmentioning
confidence: 99%
“…This procedure is described in Ref. [17], is briefly explained in Section 2.1 and need only be performed once, after the fragment meshes are generated. At a second stage, fragments are processed in pairs, in order to define a transformation that matches their fractured facets and a matching error is calculated.…”
Section: Methods Overviewmentioning
confidence: 99%
“…The surface segmentation of a fragment [17] is accomplished, using a simple region-growing algorithm. The process begins with an arbitrary polygon.…”
Section: Detection Of Fractured Sidesmentioning
confidence: 99%
“…The first one merges points that have similar region properties calculated from their neighboring points such as normal vectors [3,4] , curvatures [5] parameters of fitted planes [6,7,8] or quadratic surfaces [6,9] , and other indices corresponding to local surface shapes [10] . As calculated properties are very sensitive to noise and quantization errors, they cause over segmentation results [11] .…”
Section: Introductionmentioning
confidence: 99%