PROJECT TITLE :

Skin Detection Based On Multi-Seed Propagation In A Multi-Layer Graph For Regional And Color Consistency - 2017

ABSTRACT:

We propose a replacement skin detection methodology based mostly on multi-seeds propagation in a multi-layer graph representation of a picture. Initially, some of nodes in the graph are set to be foreground or background seeds primarily based on a simple Bayesian skin detector, and they are propagated through the graph to search out the skin probability in the way of semi-supervised learning. The graph is meant to consider not only native and international coherence however also to contemplate the colour consistency by constructing a multilayer graph of image and cluster layers. Extensive experiments on several datasets are conducted, which demonstrate that our methodology outperforms the existing methods in terms of various quantitative measures, like accuracy, precision, recall and F-live.