Please use this identifier to cite or link to this item:
https://hdl.handle.net/2440/108865
Citations | ||
Scopus | Web of Science® | Altmetric |
---|---|---|
?
|
?
|
Type: | Conference paper |
Title: | Context based re-ranking for object retrieval |
Author: | Chen, Y. Dick, A. Li, X. Hill, R. |
Citation: | Lecture Notes in Artificial Intelligence, 2015 / Cremers, D., Reid, I., Saito, H., Yang, M.H. (ed./s), vol.9003, pp.196-210 |
Publisher: | Springer International Publishing |
Issue Date: | 2015 |
Series/Report no.: | Lecture Notes in Computer Science, (LNCS, vol. 9003) |
ISBN: | 9783319168647 |
ISSN: | 0302-9743 1611-3349 |
Conference Name: | 12th Asian Conference on Computer Vision (ACCV 2014) (1 Nov 2014 - 5 Nov 2014 : Singapore) |
Editor: | Cremers, D. Reid, I. Saito, H. Yang, M.H. |
Statement of Responsibility: | Yanzhi Chen, B, Anthony Dick, Xi Li, and Rhys Hill |
Abstract: | We propose a simple but effective re-ranking method for improving the results of object retrieval. Our method considers the contextual information embedded in a dataset. This is based on the observation that if there are multiple images containing the same object in a dataset, then these images can often be grouped into clusters. We make the following two contributions. Firstly, we gain this contextual information by a random dimension partition of the dataset. This enables online query model expansion if needed. Secondly, we use the collected contextual information to refine the initial retrieval results by taking into account the context in which each retrieved image occurs. Experimental results on several datasets demonstrate the effectiveness of our method in both accuracy and computation cost: our method refines retrieval results without relying on low-level feature matching or re-issuing the query. |
Rights: | © Springer International Publishing Switzerland 2015 |
DOI: | 10.1007/978-3-319-16865-4_13 |
Published version: | http://dx.doi.org/10.1007/978-3-319-16865-4_13 |
Appears in Collections: | Aurora harvest 8 Computer Science publications |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
RA_hdl_108865.pdf Restricted Access | Restricted Access | 759.17 kB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.