Automatic Query Expansion for News Video Retrieval

Zhai, Yun; Liu, Jingen; Shah, Mubarak

doi:10.1109/icme.2006.262693

Cited by 8 publications

(4 citation statements)

References 10 publications

(7 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Within their experiment, they significantly outperform a baseline run without any query expansion, hence indicating the potentials of query modification in video search. Similar results are reported by Porkaew and Chakrabarti [90] and Zhai et al [131], who both expand search queries using content-based visual features.…”

Section: Personalised Searchsupporting

confidence: 88%

Computation for Humanity

2018

View full text Add to dashboard Cite

Recently, The Guardian, one of Britain's most popular daily newspapers published an online article 2 , recognising the fifth anniversary of the video sharing portal YouTube 3 . YouTube is at the forefront of a recent development that, in 2006, convinced the renowned Time magazine to dedicate their Person of the Year award to "You". "You" represent the millions of people that started to voluntarily generate (user-generated) content, e.g. in Wikipedia, Facebook and, of course, YouTube. More and more people do not only actively consume content, they have also started to create their own content. Thus, we are observing a paradigm change

show abstract

Section: Personalised Searchsupporting

confidence: 88%

Computation for Humanity

2018

View full text Add to dashboard Cite

show abstract

“…This technology represents semantics that videos (voices) contain by extracting vision (audition) features, such as colors, textures, shapes, tones and motions etc. And then according to retrieval examples that users submit, similar video streaming and audition streaming information are accomplished to be inspected [4]. Retrieval technology based on contents can make up the traditional retrieval technology in some way, while at present the efficient algorithms is still short to retrieve three-dimensional motion.…”

Section: Introductionmentioning

confidence: 99%

A Novel Motion Retrieval Method Based on Revised Matching Algorithm and High Dimensional Index

Xiang

2012

AMR

View full text Add to dashboard Cite

This paper proposes a method to speed up motion retrieval: matching algorithm improvement and motion high-dimensional index. The fast elimination algorithm is used to delete most of motions that are not similar with samplings. Then indexing of high-dimensional feature is done with 3D motion data structure. The experimental results show that our method outperforms the classical motion retrieval method both in speed and accuracy.

show abstract

“…Intuitively, the information content of a term reflects its importance, or specificity, and is defined as: (2) where prob(c) is the probability of a term c, which can be estimated from the term's frequency in a large corpus.…”

Section: Wordnet Similarity Measuresmentioning

confidence: 99%

“…WordNet's rich vocabulary and complex structure of relationships can be used to determine semantic concepts for words as well as links to near and distant related terms. With the emergence of audio-visual material and the desire to perform information retrieval on such resources, WordNet has become an important tool used in query expansion [2]. More recently, WordNet has become a medium used to bridge the gap between semantic and visual domains, as its concept classes are linked to visual counterparts [3].…”

Section: Introductionmentioning

confidence: 99%

Web-based information content and its application to concept-based video retrieval

Haubold

Natsev

2008

Proceedings of the 2008 International Conference on Content-Based Image and Video Retrieval

View full text Add to dashboard Cite

Semantic similarity between words or phrases is frequently used to find matching correlations between search queries and documents when straightforward matching of terms fails. This is particularly important for searching in visual databases, where pictures or video clips have been automatically tagged with a small set of semantic concepts based on analysis and classification of the visual content. Here, the textual description of documents is very limited, and semantic similarity based on WordNet's cognitive synonym structure, along with information content derived from term frequencies, can help to bridge the gap between an arbitrary textual query and a limited vocabulary of visual concepts. This approach, termed concept-based retrieval, has received significant attention over the last few years, and its success is highly dependent on the quality of the similarity measure used to map textual query terms to visual concepts.In this paper, we consider some issues of semantic similarity measures based on Information Content (IC), and propose a way to improve them. In particular, we note that most IC-based similarity measures are derived from a small and relatively outdated corpus (the Brown corpus), which does not adequately capture the usage pattern of many contemporary terms: for example, out of more than 150,000 WordNet terms, only about 36,000 are represented. This shortcoming reflects very negatively on the coverage of typical search query terms. We therefore suggest using alternative IC corpora that are larger and better aligned with the usage of modern vocabulary. We experimentally derive two such corpora using the WWW Google search engine, and show that they provide better coverage of vocabulary, while showing comparable frequencies for Brown corpus terms. Finally, we evaluate the two proposed IC corpora in the context of a concept-based video retrieval application using the TRECVID 2005, 2006, and 2007 datasets, and we show that they increase average precision results by up to 200%.

show abstract

Automatic Query Expansion for News Video Retrieval

Cited by 8 publications

References 10 publications

Computation for Humanity

Computation for Humanity

A Novel Motion Retrieval Method Based on Revised Matching Algorithm and High Dimensional Index

Web-based information content and its application to concept-based video retrieval

Contact Info

Product

Resources

About