Density-based indexing for approximate nearest-neighbor queries
KDD '99 Proceedings of the fifth ACM SIGKDD international conference on Knowledge discovery and data mining
SIMPLIcity: Semantics-Sensitive Integrated Matching for Picture LIbraries
IEEE Transactions on Pattern Analysis and Machine Intelligence
Clustering for Approximate Similarity Search in High-Dimensional Spaces
IEEE Transactions on Knowledge and Data Engineering
VLDB '98 Proceedings of the 24rd International Conference on Very Large Data Bases
Contorting high dimensional data for efficient main memory KNN processing
Proceedings of the 2003 ACM SIGMOD international conference on Management of data
IEEE Transactions on Knowledge and Data Engineering
SS-ClusterTree: a subspace clustering based indexing algorithm over high-dimensional image features
CIVR '08 Proceedings of the 2008 international conference on Content-based image and video retrieval
A kurtosis-based dynamic approach to Gaussian mixture modeling
IEEE Transactions on Systems, Man, and Cybernetics, Part A: Systems and Humans
Hi-index | 0.00 |
This paper proposes a novel clustering based indexing approach called GMM-ClusterForest for supporting multi-features based similarity search in high-dimensional spaces. We fit a Gaussian Mixture Model (GMM) to data through the Expectation-Maximization (EM) algorithm for estimating GMM parameters and the Minimum Description Length (MDL) criterion for selecting GMM structure. Each Gaussian component in the GMM is taken as a cluster center and each data point is assigned to the cluster according to the Bayesian decision rule. By performing this clustering method hierarchically, an index tree is constructed and the corresponding similarity search method is developed for a type of features. Then multi-features based similarity search is fulfilled by fusing the index trees for all the types of features considered. We evaluated the proposed indexing approach through applying it to example-based image retrieval and conducting the experiments on Corel 1000 dataset and self-collected large dataset. The experimental results show that our approach is effective and promising.