2017
DOI: 10.1109/tpami.2016.2545667
|View full text |Cite
|
Sign up to set email alerts
|

Higher-Order Occurrence Pooling for Bags-of-Words: Visual Concept Detection

Abstract: Abstract-In object recognition, the Bag-of-Words model assumes: i) extraction of local descriptors from images, ii) embedding the descriptors by a coder to a given visual vocabulary space which results in mid-level features, iii) extracting statistics from mid-level features with a pooling operator that aggregates occurrences of visual words in images into signatures, which we refer to as First-order Occurrence Pooling. This paper investigates higher-order pooling that aggregates over co-occurrences of visual … Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
156
0
3

Year Published

2017
2017
2020
2020

Publication Types

Select...
5
3
1

Relationship

3
6

Authors

Journals

citations
Cited by 103 publications
(159 citation statements)
references
References 48 publications
0
156
0
3
Order By: Relevance
“…For example, visual phrases [103], [104], [105], [106] are generated among individual visual words to provide more strict matching criterion. Visual word cooccurrences in the entire image are estimated [107] and aggregated [108], while in [109], [110], [29] visual word clusters within local neighborhoods are discovered. Visual phrases can also be constructed from adjacent image patches [103], random spatial partitioning [106], and localized stable regions [29] such as MSER [28].…”
Section: Geometric Matchingmentioning
confidence: 99%
“…For example, visual phrases [103], [104], [105], [106] are generated among individual visual words to provide more strict matching criterion. Visual word cooccurrences in the entire image are estimated [107] and aggregated [108], while in [109], [110], [29] visual word clusters within local neighborhoods are discovered. Visual phrases can also be constructed from adjacent image patches [103], random spatial partitioning [106], and localized stable regions [29] such as MSER [28].…”
Section: Geometric Matchingmentioning
confidence: 99%
“…В данной работе в качестве метода представления данных был выбран Мешок слов (Bag of Words, BoW) [11]. Несмотря на то, что этот метод первоначально был создан для представления текстовых данных, он также был успешно использован в работах [7,8,12] для представления визуальных данных.…”
Section: метод представления данных: Bag Of Wordsunclassified
“…It is somehow satisfied with the classical inner product k(x, y) = x|y . Several authors [12,34,36,56] propose to increase the contrast between related and unrelated features with a monomial match kernel of degree p of the form…”
Section: Background: Match Kernels and Embeddingsmentioning
confidence: 99%