Please use this identifier to cite or link to this item: http://hdl.handle.net/10397/15099
Title: Exploring spatial correlation for visual object retrieval
Authors: Shi, M
Sun, X
Tao, D
Xu, C
George, B 
Liu, H
Keywords: BOVW
Spatial correlation
High-order predictor
Penalty tree
Co-Cosine
Co-TFIDF
Issue Date: 2015
Publisher: Association for Computing Machinary
Source: ACM transactions on intelligent systems and technology, 2015, p. 1-21 How to cite?
Journal: ACM transactions on intelligent systems and technology 
Abstract: Bag-of-visual-words (BOVW) based image representation has received intense attention in recent years and has improved content based image retrieval (CBIR) significantly. BOVW does not consider the spatial correlation between visual words in natural images, and thus, biases the generated visual words towards noise when the corresponding visual features are not stable. This paper outlines the construction of a visual word co-occurrence matrix by exploring visual word co-occurrence extracted from small affine-invariant regions in a large collection of natural images. Based on this co-occurrence matrix, we first present a novel high-order predictor to accelerate the generation of spatially correlated visual words, and a penalty tree (PTree) to continue generating the words after the prediction. Subsequently, we propose two methods of co-occurrence weighting similarity measure for image ranking: Co-Cosine and Co-TFIDF. These two new schemes down-weight the contributions of the words that are less discriminative because of frequent cooccurrences with other words. We conduct experiments on Oxford and Paris Building datasets, in which the ImageNet dataset is used to implement a large scale evaluation. Cross dataset evaluations between Oxford and Paris datasets, Oxford and Holidays datasets are also provided. Thorough experimental results suggest that our method outperforms the state-of-the-art without adding much additional cost to the BOVW model.
URI: http://hdl.handle.net/10397/15099
ISSN: 2157-6904
EISSN: 2157-6912
Appears in Collections:Journal/Magazine Article

Access
View full-text via PolyU eLinks SFX Query
Show full item record

Page view(s)

60
Last Week
0
Last month
Checked on Aug 14, 2017

Google ScholarTM

Check



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.