2008
DOI: 10.1016/j.cviu.2008.05.009
|View full text |Cite
|
Sign up to set email alerts
|

A general shape context framework for object identification

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
5
0
1

Year Published

2011
2011
2022
2022

Publication Types

Select...
5
2
1

Relationship

0
8

Authors

Journals

citations
Cited by 11 publications
(6 citation statements)
references
References 31 publications
0
5
0
1
Order By: Relevance
“…the intersection needs to produce one (or more) entries of Ĩ . The filter procedure is essential to ensure efficiency especially when the feature sets and database are large [15]. Finally, each neighborhood-match votes on the optimal model.…”
Section: Similarity Indexingmentioning
confidence: 99%
See 1 more Smart Citation
“…the intersection needs to produce one (or more) entries of Ĩ . The filter procedure is essential to ensure efficiency especially when the feature sets and database are large [15]. Finally, each neighborhood-match votes on the optimal model.…”
Section: Similarity Indexingmentioning
confidence: 99%
“…Multiple-phase encoded texture features are proposed to capture the characters of skeptical regions spatially and chronologically. For tumor classification, hyper-cube structure based indexing [15,16] is used to query an image database of tumor sample models. The similarity score is calculated to rank the retrieval results.…”
Section: Introductionmentioning
confidence: 99%
“…The proposed shape features are invariant to scale changes and rotation. Pair-wise geometric attributes (PGAs), proposed by Chi et al [5,28], capture the geometric relation between a pair of circular arcs of lines; they are among the most recent shape features of circular arcs. Compared to PGAs, our binary shape descriptor is advantageous in the following aspects.…”
Section: Binary Arc Descriptormentioning
confidence: 99%
“…Circular arcs have been used for object recognition [23] and image retrieval [5, 24] in previous studies. Our objective is to use circular arcs for object detection in real images, which has been rarely attempted so far.…”
Section: Introductionmentioning
confidence: 99%
“…Para esta dissertação, é explorado apenas a representação por região, pois os conjuntos contém imagens em que a informação de textura e cor são mais salientes. Entretanto, é possível empregar outras técnicas para descrever imagens, tais como as baseadas em forma dos objetos (Chi and Leung (2008); Datta et al (2008); Costa and Cesar Jr. (2009)). Na Seção 3.3 foram descritas duas técnicas tradicionais no contexto de textura, matriz de co-ocorrência e filtros de…”
Section: Considerações Finaisunclassified