Abstract | ||
---|---|---|
We introduce a new graph cut for clustering which we call the Information Cut. It is derived using Parzen windowing to estimate an information theoretic distance measure between probability density functions. We propose to optimize the Information Cut using a gradient descent-based approach. Our algorithm has several advantages compared to many other graph-based methods in terms of determining an appropriate affinity measure, computational complexity, memory requirements and coping with different data scales. We show that our method may produce clustering and image segmentation results comparable or better than the state-of-the art graph-based methods. |
Year | DOI | Venue |
---|---|---|
2007 | 10.1016/j.patcog.2006.06.028 | Pattern Recognition |
Keywords | Field | DocType |
gradient descent approach,new graph cut,information theory,graph theoretic cut,image segmentation result,appropriate affinity measure,information cut,annealing.,graph-based method,different data scale,memory requirement,gradient descent optimization,gradient descent-based approach,parzen window density estimation,information theoretic distance measure,computational complexity,annealing,clustering,gradient descent,graph cut,probability density function,density estimation,image segmentation | Gradient method,Graph theory,Density estimation,Cut,Gradient descent,Pattern recognition,Algorithm,Image segmentation,Artificial intelligence,Cluster analysis,Mathematics,Computational complexity theory | Journal |
Volume | Issue | ISSN |
40 | 3 | Pattern Recognition |
Citations | PageRank | References |
15 | 0.68 | 19 |
Authors | ||
5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Robert Jenssen | 1 | 370 | 43.06 |
Deniz Erdogmus | 2 | 1299 | 169.92 |
Kenneth E. Hild, II | 3 | 71 | 6.58 |
Jose C. Principe | 4 | 2295 | 282.29 |
Torbjørn Eltoft | 5 | 583 | 48.56 |