2012 IEEE Conference on Computer Vision and Pattern Recognition (2012)
Providence, RI USA
June 16, 2012 to June 21, 2012
K-means, a simple and effective clustering algorithm, is one of the most widely used algorithms in computer vision community. Traditional k-means is an iterative algorithm - in each iteration new cluster centers are computed and each data point is re-assigned to its nearest center. The cluster re-assignment step becomes prohibitively expensive when the number of data points and cluster centers are large. In this paper, we propose a novel approximate k-means algorithm to greatly reduce the computational complexity in the assignment step. Our approach is motivated by the observation that most active points changing their cluster assignments at each iteration are located on or near cluster boundaries. The idea is to efficiently identify those active points by pre-assembling the data into groups of neighboring points using multiple random spatial partition trees, and to use the neighborhood information to construct a closure for each cluster, in such a way only a small number of cluster candidates need to be considered when assigning a data point to its nearest cluster. Using complexity analysis, real data clustering, and applications to image retrieval, we show that our approach out-performs state-of-the-art approximate k-means algorithms in terms of clustering quality and efficiency.
trees (mathematics), approximation theory, iterative methods, pattern clustering, clustering efficiency, approximate k-means clustering algorithm, cluster closures, computer vision, iterative algorithm, cluster reassignment step, multiple random spatial partition trees, complexity analysis, real data clustering, clustering quality, Clustering algorithms, Vegetation, Image retrieval, Visualization, Complexity theory, Algorithm design and analysis, Instruction sets
Shipeng Li, Gang Zeng, Qifa Ke, Jingdong Wang and Jing Wang, "Fast approximate k-means via cluster closures," 2012 IEEE Conference on Computer Vision and Pattern Recognition(CVPR), Providence, RI USA, 2012, pp. 3037-3044.