Agrupamiento por similitud de imágenes mediante árbol de expansión mínima y soft heap

Abstract

This research proposes a grouping of faces using a minimum spanning tree. The Prim’s algorithm is optimized using the structure Soft Heap as priority queue, and tests are done to check its efficiency compared to Prim’s algorithm with a Fibonacci Heap as queue priority. The characteristic vector is obtained by the Haar wavelet, and for the distances between images are used Euclidean distance between their feature vectors.

Download paper here

BibTeX:

@InProceedings{LFR12,
    author       = { {Lopez del Alamo}, Cristian J. and {Fuentes Perez}, Lizeth J. and {Romero Calla}, Luciano A. },
    title        = { Agrupamiento por Similitud de Imágenes mediante árbol de Expansión Mínima y Soft Heap },
    booktitle    = { Encuentro Chileno de Computación, Jornadas Chilenas de Computación },
    year         = { 2012 }
}