A comprehensive image similarity retrieval system that utilizes multiple feature vectors in high dimensional space

Curtis, K.M.; Taniguchi, Noburou; Nakagawa, Junya; Yamamuro, Masashi

doi:10.1109/icics.1997.647083

Cited by 6 publications

(2 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The vectors are individually indexed according to feature and entered in the database's server. Indices are made based on an improved version of VAMSplit R-tree [2], which provides faster retrieval.…”

Section: Database Constructionmentioning

confidence: 99%

See 1 more Smart Citation

A practical query-by-humming system for a large music database

Kosugi

Nishihara

Sakata

et al. 2000

Proceedings of the Eighth ACM International Conference on Multimedia

View full text Add to dashboard Cite

A music retrieval system that accepts hummed tunes as queries is described in this paper. This system uses similaxity retrieval because a hummed tune may contain errors. The retrieval result is a list of song names ranked according to the closeness of the match. Our ultimate goal is that the correct song should be first on the list. This means that eventually our system's similarity retrieval should allow for only one correct answer.The most significant improvement our system has over general query-by-humming systems is that all processing of musical information is done based on beats instead of notes. This type of query processing is robust against queries generated from erroneous input. In addition, acoustic information is transcribed and converted into relative intervals and is used for making feature vectors. This increases the resolution of the retrieval system compared with other general systems, which use only pitch direction information.The database currently holds over 10,000 songs, and the retrieval time is at most one second. This level of performance is mainly achieved through the use of indices for retrieval. In this paper, we also report on the results of music analyses of the songs in the database. Based on these results, new technologies for improving retrieval accuracy, such as partial feature vectors and or'ed retrieval among multiple search keys, axe proposed. The effectiveness of these technologies is evaluated quantitatively, and it is found that the retrieval accuracy increases by more than 20% compared with the previous system [9]. Practical user interfaces for the system axe also described.

show abstract

Section: Database Constructionmentioning

confidence: 99%

“…The authors have developed the HyperMatch engine [2], a high-speed similarity retrieval engine based on distance measurement employing multiple high-dimensional feature vectors. Each vector retains feature information, and similaxity is represented as a weighted combination of the search results from individual feature vector spaces.…”

mentioning

confidence: 99%

A practical query-by-humming system for a large music database

Kosugi

Nishihara

Sakata

et al. 2000

Proceedings of the Eighth ACM International Conference on Multimedia

View full text Add to dashboard Cite

show abstract

ExSight: Highly Accurate Object Based Image Retrieval System Enhanced by Redundant Object Extraction

Kushima¹,

Akama²,

Kon'ya³

et al. 2000

Web-Age Information Management

View full text Add to dashboard Cite

Hierarchical Space Model for Multimedia Data Retrieval

Onizuka¹,

Nishioka²

2000

Advances in Visual Information Management

View full text Add to dashboard Cite

Database management systems for multimedia data retrieval are becoming more important, as digital videos and cameras increase in popularity.An important feature of multimedia data retrieval is that users rarely specify their first queries exactly, and must clarify what they want by browsing the query results, refining their query by trial and error. It is therefore desirable for a multimedia database management system (DBMS) to develop a rough query result quickly and refine it over time. This paper describes a hierarchical space model for the multimedia data retrieval that is similar to that of the human memory hierarchy. The aim of the hierarchical space model is to improve the similarity retrieval's performance with little loss in query result quality. We implemented the hierarchical space model on the ORDBMS LiteObject and applied it to an image retrieval application. The results of this test proved the efficiency of our hierarchical space model.

show abstract

A comprehensive image similarity retrieval system that utilizes multiple feature vectors in high dimensional space

Cited by 6 publications

References 3 publications

A practical query-by-humming system for a large music database

A practical query-by-humming system for a large music database

ExSight: Highly Accurate Object Based Image Retrieval System Enhanced by Redundant Object Extraction

Hierarchical Space Model for Multimedia Data Retrieval

Contact Info

Product

Resources

About