Abstract | ||
---|---|---|
A number of minimum spanning tree algorithms have been proposed for lossy compression of image sets. In these algorithms, a complete graph is constructed from the entire image set and possibly an average image, and a minimum spanning tree is used to determine which difference images to encode. In this paper, we pro- pose a hierarchical minimum spanning tree algorithm in which the minimum spanning tree algorithm is first applied to clusters of similar images and then it is ap- plied to the average images of the clusters. It is shown that the new algorithm outperforms the previous image set compression algorithms for image sets which are not very similar, especially at lower bitrates. Furthermore, the computational requirement for a hierarchical mini- mum spanning tree is significantly lower than the pre- vious minimum spanning tree algorithms when the cost of clustering can be neglected. |
Year | Venue | Keywords |
---|---|---|
2008 | IPCV | image set compression,minimum spanning tree.,clustering,spanning tree,lossy compression,minimum spanning tree,compression algorithm,complete graph |
Field | DocType | Citations |
Discrete mathematics,Distributed minimum spanning tree,Computer science,Prim's algorithm,Algorithm,Euclidean minimum spanning tree,Spanning tree,Minimum spanning tree-based segmentation,Kruskal's algorithm,Reverse-delete algorithm,Minimum spanning tree | Conference | 2 |
PageRank | References | Authors |
0.44 | 8 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Anthony Schmieder | 1 | 14 | 1.49 |
Howard Cheng | 2 | 137 | 18.85 |
Barry Gergel | 3 | 25 | 2.84 |
Xiaobo Li | 4 | 502 | 37.50 |