Document Expansion for Text-Based Image Retrieval at CLEF 2009

Min, Jinming; Wilkins, Peter; Leveling, Johannes; Jones, Gareth J. F.

doi:10.1007/978-3-642-15751-6_19

Cited by 7 publications

(11 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Furthermore, the best performing group in 2009, deuceng (Kilinc and Alpkocak, 2009), also applied document expansion using semantic knowledge bases. Document and query expansion using DBpedia were also applied by dcu (Min et al, 2010) in 2009 and achieved improvements over their textual baseline. All this constitutes strong evidence that such expansion techniques, particularly when applied judiciously so as to deal with the noise that can be potentially added, are particularly effective for such collections of images that are accompanied by short and possibly noisy textual descriptions.…”

Section: Best Practicesmentioning

confidence: 99%

“…For further details, see Chapter 14 in this volume. dcu (2009) (Min et al, 2010) They focused their experimentations on the expansion of the images' textual descriptions and of the textual part of the topics, using the Wikipedia abstracts' collection DBpedia 8 and blind relevance feedback. When DBpedia was employed, the terms from its top-ranked documents retrieved in response to the image description (or textual query) were sorted by their frequency and the top-ranked were selected to expand the images' (or queries') text.…”

Section: Cea (2008 2009)mentioning

confidence: 99%

See 1 more Smart Citation

The Wikipedia Image Retrieval Task

Tsikrika

Kludas

2010

The Information Retrieval Series

View full text Add to dashboard Cite

The Wikipedia image retrieval task at ImageCLEF provides a test-bed for the system-oriented evaluation of visual information retrieval from a collection of Wikipedia images. The aim is to investigate the effectiveness of retrieval approaches that exploit textual and visual evidence in the context of a large and heterogeneous collection of images that are searched for by users with diverse information needs. This chapter presents an overview of the available test collections, summarises the retrieval approaches employed by the groups that participated in the task during the 2008 and 2009 ImageCLEF campaigns, provides an analysis of the main evaluation results, identifies best practices for effective retrieval, and discusses open issues.

show abstract

Section: Best Practicesmentioning

confidence: 99%

Section: Cea (2008 2009)mentioning

confidence: 99%

The Wikipedia Image Retrieval Task

Tsikrika

Kludas

2010

The Information Retrieval Series

View full text Add to dashboard Cite

show abstract

“…image tags) as output [2]. The system integrates methods to reduce and expand the image tag set, thus decreasing the effect of sparse image tags.…”

Section: Extended Abstractmentioning

confidence: 99%

“…The system integrates methods to reduce and expand the image tag set, thus decreasing the effect of sparse image tags. It builds on different image collections such as the Wikipedia image collection 1 and the Microsoft Office.com ClipArt collection 2 , but can be applied to social collections such as Flickr as well. Our demonstration system runs on PCs, tablets, and smartphones, making use of advanced user interface capabilities on mobile devices.…”

Section: Extended Abstractmentioning

confidence: 99%

Multi-platform image search using tag enrichment

Min

Lopes

Leveling

et al. 2012

Proceedings of the 35th International ACM SIGIR Conference on Research and Development in Information Retrieval

Self Cite

View full text Add to dashboard Cite

EXTENDED ABSTRACTThe number of images available online is growing steadily and current web search engines have indexed more than 10 billion images. Approaches to image retrieval are still often text-based and operate on image annotations and captions. Image annotations (i.e. image tags) are typically short, user-generated, and of varying quality, which increases the mismatch problem between query terms and image tags. For example, a user might enter the query wedding dress while all images are annotated with bridal gown or wedding gown. This demonstration presents an image search system using reduction and expansion of image annotations to overcome vocabulary mismatch problems by enriching the sparse set of image tags.Our image search application accepts a written query as input and produces a ranked list of result images and annotations (i.e. image tags) as output [2]. The system integrates methods to reduce and expand the image tag set, thus decreasing the effect of sparse image tags. It builds on different image collections such as the Wikipedia image collection 1 and the Microsoft Office.com ClipArt collection 2 , but can be applied to social collections such as Flickr as well. Our demonstration system runs on PCs, tablets, and smartphones, making use of advanced user interface capabilities on mobile devices. Figure 1 presents an overview of the complete system for Document Expansion (DE) using Wikipedia as an external resource. Important terms in the original image annotation are used as a query to retrieve Wikipedia articles and extract DE terms from the topranked documents. These terms are added to the original image tags to form the enriched tag set. The expanded annotations are indexed in the retrieval system. The key stage here is to select key terms from the documents prior to expansion in a process we refer to as Document Reduction (DR). The objective of DR is to focus the DE "query" on the most important elements in the tag set.The system is implemented as a client-server architecture, where the client realizes the graphical user interface and handles query 1 http://www.imageclef.org/wikidata 2

show abstract

“…Hersh et al [10] expand the query from web pages online in a genomic IR task. Our own previous work [11] reports initial experiments on QE from Wikipedia for a text-based image retrieval tasks, and shows improvement compared with the QE from the target corpus. We extend this earlier work in this paper.…”

Section: Background and Related Workmentioning

confidence: 99%

External Query Reformulation for Text-Based Image Retrieval

Min

Jones

2011

String Processing and Information Retrieval

Self Cite

View full text Add to dashboard Cite

Abstract. In text-based image retrieval, the Incomplete Annotation Problem (IAP) can greatly degrade retrieval effectiveness. A standard method used to address this problem is pseudo relevance feedback (PRF) which updates user queries by adding feedback terms selected automatically from top ranked documents in a prior retrieval run. PRF assumes that the target collection provides enough feedback information to select effective expansion terms. This is often not the case in image retrieval since images often only have short metadata annotations leading to the IAP. Our work proposes the use of an external knowledge resource (Wikipedia) in the process of refining user queries. In our method, Wikipedia documents strongly related to the terms in user query ("definition documents") are first identified by title matching between the query and titles of Wikipedia articles. These definition documents are used as indicators to re-weight the feedback documents from an initial search run on a Wikipedia abstract collection using the Jaccard coefficient. The new weights of the feedback documents are combined with the scores rated by different indicators. Query-expansion terms are then selected based on these new weights for the feedback documents. Our method is evaluated on the ImageCLEF WikipediaMM image retrieval task using text-based retrieval on the document metadata fields. The results show significant improvement compared to standard PRF methods.

show abstract

Document Expansion for Text-Based Image Retrieval at CLEF 2009

Cited by 7 publications

References 2 publications

The Wikipedia Image Retrieval Task

The Wikipedia Image Retrieval Task

Multi-platform image search using tag enrichment

External Query Reformulation for Text-Based Image Retrieval

Contact Info

Product

Resources

About