Spatial indexing of high-dimensional data based on relative approximation

Sakurai, Yasushi; Yoshikawa, Masatoshi; Uemura, Shunsuke; Kojima, Hironori

doi:10.1007/s00778-002-0066-9

Cited by 27 publications

(16 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Several indexing structures [2,4,10,11,12,13] have been proposed. However, all these techniques are for indexing an individual feature space purpose, and they all suffer from known "dimensionality curse".…”

Section: Related Workmentioning

confidence: 99%

Indexing Text and Visual Features for WWW Images

Shen

Zhou

Cui

2005

Web Technologies Research and Development - APWeb 2005

View full text Add to dashboard Cite

In this paper, we present a novel indexing technique called Multi-scale Similarity Indexing (MSI) to index image's multi-features into a single one-dimensional structure. Both for text and visual feature spaces, the similarity between a point and a local partition's center in individual space is used as the indexing key, where similarity values in different features are distinguished by different scale. Then a single indexing tree can be built on these keys. Based on the property that relevant images haves similar similarity values from the center of the same local partition in any feature space, certain number of irrelevant images can be fast pruned based on the triangle inequity on indexing keys. To remove the "dimensionality curse" existing in high dimensional structure, we propose a new technique called Local Bit Stream (LBS). LBS transforms image's text and visual feature representations into simple, uniform and effective bit stream (BS) representations based on local partition's center. Such BS representations are small in size and fast for comparison since only bit operation are involved. By comparing common bits existing in two BSs, most of irrelevant images can be immediately filtered. Our extensive experiment showed that single one-dimensional index on multi-features improves multi-indices on multi-features greatly. Our LBS method outperforms sequential scan on high dimensional space by an order of magnitude.

show abstract

Section: Related Workmentioning

confidence: 99%

Indexing Text and Visual Features for WWW Images

Shen

Zhou

Cui

2005

Web Technologies Research and Development - APWeb 2005

View full text Add to dashboard Cite

show abstract

“…Most existing hierarchical index structures (e.g., R-tree [15], R*-tree [3], and A-tree [31]) were not designed specifically for target search, which typically cannot be answered in one iteration and may require auxiliary information (e.g., sampling points) to answer sampling queries and constrained sampling queries. Collecting auxiliary information on the fly during each feedback iteration causes overheads on CPU and disk I/O.…”

Section: Related Workmentioning

confidence: 99%

Fast Query Point Movement Techniques for Large CBIR Systems

Liu

Hua

et al. 2009

IEEE Trans. Knowl. Data Eng.

View full text Add to dashboard Cite

Abstract-Target search in content-based image retrieval (CBIR) systems refers to finding a specific (target) image such as a particular registered logo or a specific historical photograph. Existing techniques, designed around query refinement based on relevance feedback, suffer from slow convergence, and do not guarantee to find intended targets. To address these limitations, we propose several efficient query point movement methods. We prove that our approach is able to reach any given target image with fewer iterations in the worst and average cases. We propose a new index structure and query processing technique to improve retrieval effectiveness and efficiency. We also consider strategies to minimize the effects of users' inaccurate relevance feedback. Extensive experiments in simulated and realistic environments show that our approach significantly reduces the number of required iterations and improves overall retrieval performance. The experimental results also confirm that our approach can always retrieve intended targets even with poor selection of initial query points.Index Terms-Content-based image retrieval, relevance feedback, target search, index structures.

show abstract

“…The reader can refer to [5], [11], [12], [13] for comprehensive reviews of existing nearest neighbor methods. A large amount of work focuses on efficient nearest neighbor retrieval in multidimensional vector spaces [14], [15], [16], [17], [18], [19], [20], [21], [22]. Particular mention should be made to Locality Sensitive Hashing (LSH) [23], an approximate nearest neighbor method that has been shown theoretically to scale well with the number of dimensions and has produced good results in practice [24], [25], [26].…”

Section: Related Workmentioning

confidence: 99%

BoostMap: An Embedding Method for Efficient Nearest Neighbor Retrieval

Athitsos

Alon

Sclaroff

et al. 2008

IEEE Trans. Pattern Anal. Mach. Intell.

View full text Add to dashboard Cite

Abstract-This paper describes BoostMap, a method for efficient nearest neighbor retrieval under computationally expensive distance measures. Database and query objects are embedded into a vector space in which distances can be measured efficiently. Each embedding is treated as a classifier that predicts for any three objects X, A, B whether X is closer to A or to B. It is shown that a linear combination of such embedding-based classifiers naturally corresponds to an embedding and a distance measure. Based on this property, the BoostMap method reduces the problem of embedding construction to the classical boosting problem of combining many weak classifiers into an optimized strong classifier. The classification accuracy of the resulting strong classifier is a direct measure of the amount of nearest neighbor structure preserved by the embedding. An important property of BoostMap is that the embedding optimization criterion is equally valid in both metric and nonmetric spaces. Performance is evaluated in databases of hand images, handwritten digits, and time series. In all cases, BoostMap significantly improves retrieval efficiency with small losses in accuracy compared to brute-force search. Moreover, BoostMap significantly outperforms existing nearest neighbor retrieval methods such as Lipschitz embeddings, FastMap, and VP-trees.

show abstract

Spatial indexing of high-dimensional data based on relative approximation

Cited by 27 publications

References 15 publications

Indexing Text and Visual Features for WWW Images

Indexing Text and Visual Features for WWW Images

Fast Query Point Movement Techniques for Large CBIR Systems

BoostMap: An Embedding Method for Efficient Nearest Neighbor Retrieval

Contact Info

Product

Resources

About