Mining near duplicate image groups
Jing Li1; Xueming Qian1; Qing Li1; Yisi Zhao1; Liejun Wang2; Yuan Yan Tang3
Source PublicationMultimedia Tools and Applications

Most recently the social media sharing websites such as Flickr, Facebook, and Picasa have allowed users to share their personal photos with friends. Moreover, people like to follow, forward their favorite images, which is one of the main source of near duplicate images. And also, the worldwide place of interests such as Roma, Statue of Liberty and London Tower Bridge etc., attract world-wide visitors. For these places, travelers take photos, write travelogues and share them with their social friends. The photos taken from the same place with or without viewpoint variations are near duplicate images. How to detect them is an ad-hoc problem in the area of image analysis and multimedia processing. The existing near duplicate image processing approaches mainly focused on finding the near duplicate images for a given input image, where a query image is needed. However, how to find the near duplicate image groups (NDIG) automatically from the web-scale social images is very challenging. So, in this paper, instead of searching near duplicates image for certain input image, we proposed an automatic NDIG mining approach by utilizing adaptive global feature clustering and local feature refinement. The proposed NDIG mining approach is achieved by utilizing a hierarchical model. It is a two-layer hierarchical structure by first utilizing adaptive global feature clustering based candidate NDIG detection and then using local feature refinement based NDIG verification. The global clustering is mainly for reducing computational cost for processing the large scale image set. The local refinement is for improving NDIG detection performances. Experiments on four datasets show the effectiveness of our approach.

KeywordImage Retrieval Near Duplicate Image Group Social Media
URLView the original
Indexed BySCI
WOS Research AreaComputer Science ; Engineering
WOS SubjectComputer Science, Information Systems ; Computer Science, Software Engineering ; Computer Science, Theory & Methods ; Engineering, Electrical & Electronic
WOS IDWOS:000348445300019
Fulltext Access
Citation statistics
Cited Times [WOS]:16   [WOS Record]     [Related Records in WOS]
Document TypeJournal article
CollectionUniversity of Macau
Corresponding AuthorXueming Qian; Liejun Wang
Affiliation1.SMILES LAB at School of Electronics and Information Engineering, Xi’an Jiaotong University, Xi’an, China
2.Xinjiang University, Urumqi, China
3.Macau University, Macau, China
Recommended Citation
GB/T 7714
Jing Li,Xueming Qian,Qing Li,et al. Mining near duplicate image groups[J]. Multimedia Tools and Applications,2015,74(2):655-669.
APA Jing Li,Xueming Qian,Qing Li,Yisi Zhao,Liejun Wang,&Yuan Yan Tang.(2015).Mining near duplicate image groups.Multimedia Tools and Applications,74(2),655-669.
MLA Jing Li,et al."Mining near duplicate image groups".Multimedia Tools and Applications 74.2(2015):655-669.
Files in This Item:
There are no files associated with this item.
Related Services
Recommend this item
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Jing Li]'s Articles
[Xueming Qian]'s Articles
[Qing Li]'s Articles
Baidu academic
Similar articles in Baidu academic
[Jing Li]'s Articles
[Xueming Qian]'s Articles
[Qing Li]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Jing Li]'s Articles
[Xueming Qian]'s Articles
[Qing Li]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.

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