Difference between revisions of "Image interpretation using large corpus: Wikipedia"

Revision as of 18:42, October 18, 2013 (view source) Ochado (Talk | contribs) m (Text replace - "([ ][|]research_design=)([^ ]*Experiment)([^ ]*[ ][|]collected_datatype=)([^ ]*)([^ ]*[ ])" to "\1\2\3Experiment responses, \4\5") ← Older edit		Revision as of 18:25, December 3, 2013 (view source) Mehdi (Talk | contribs) (changed Collected data time dimension) Newer edit →
Line 31:		Line 31:
	|research_design=Experiment		|research_design=Experiment
	|collected_datatype=Experiment responses, Wikipedia pages		|collected_datatype=Experiment responses, Wikipedia pages
−	|collected_data_time_dimension=N/A	+	|collected_data_time_dimension=Cross-sectional
	|unit_of_analysis=Article		|unit_of_analysis=Article
	|wikipedia_data_extraction=Live Wikipedia		|wikipedia_data_extraction=Live Wikipedia

---

Revision as of 18:25, December 3, 2013

Publication (help)
Image interpretation using large corpus: Wikipedia
Authors:	Mandar Rahurkar, Shen-Fu Tsai, Charlie Dagli, Thomas S. Huang [edit item]
Citation:	Proceedings of the IEEE 98 (8): 1509-25. 2010.
Publication type:	Journal article
Peer-reviewed:	Yes
Database(s):
DOI:	10.1109/JPROC.2010.2050410.
Google Scholar cites:	Citations
Link(s):	Paper link
Added by Wikilit team:	Added on initial load
Search
Article:	Google Scholar BASE PubMed
Other scholarly wikis:	AcaWiki Brede Wiki WikiPapers
Web search:	Bing Google Yahoo! — Google PDF
Other:
Services
Format:	BibTeX

Image interpretation using large corpus: Wikipedia is a publication by Mandar Rahurkar, Shen-Fu Tsai, Charlie Dagli, Thomas S. Huang.

[edit] Abstract

Image is a powerful medium for expressing one's ideas and rightly confirms the adage, One picture is worth a thousand words. In this work, we explore the application of world knowledge in the form of Wikipedia to achieve this objective-literally. In the first part, we disambiguate and rank semantic concepts associated with ambiguous keywords by exploiting link structure of articles in Wikipedia. In the second part, we explore an image representation in terms of keywords which reflect the semantic content of an image. Our approach is inspired by the desire to augment low-level image representation with massive amounts of world knowledge, to facilitate computer vision tasks like image retrieval based on this information. We represent an image as a weighted mixture of a predetermined set of concrete concepts whose definition has been agreed upon by a wide variety of audience. To achieve this objective, we use concepts defined by Wikipedia articles, e.g., sky, building, or automobile. An important advantage of our approach is availability of vast amounts of highly organized human knowledge in Wikipedia. Wikipedia evolves rapidly steadily increasing its breadth and depth over time.

[edit] Research questions

"In this work, we propose a unifying application that semantically sorts images with respect to the input text query. It is composed of keyword disambiguation and image-to-semantic-concept mapping. By mining structure and statistic of links among articles in Wikipedia, keywords that are article titles are disambiguated and ranked. As for image-to-semantic-concept mapping, after bag-ofwords image feature extraction [2], it is obtained by mining captions and locations of images in Wikipedia."

Research details

Topics:	Multimedia information retrieval [edit item]
Domains:	Computer science [edit item]
Theory type:	Design and action [edit item]
Wikipedia coverage:	Sample data [edit item]
Theories:	"Undetermined" [edit item]
Research design:	Experiment [edit item]
Data source:	[edit item]
Collected data time dimension:	Cross-sectional [edit item]
Unit of analysis:	Article [edit item]
Wikipedia data extraction:	Live Wikipedia [edit item]
Wikipedia page type:	Article [edit item]
Wikipedia language:	Not specified [edit item]

[edit] Conclusion

"In this work, we proposed, implemented, and evaluated an algorithmfor expressing images as weighted representation of cognitive concepts to facilitate image understanding. We leveraged crowd sourced encyclopedia Wikipedia to learn basis representation for these concepts. Two fundamental algorithms were described: concept disambiguation and reranking that disambiguates and clarifies keywords, and image-to-concept mapping which acts as a bridge between visual feature and textual representation. These two parts are integral to image ranking systemwhich ranks images based on their semantic relevance to query keywords."

[edit] Comments

Further notes[edit]