Abstract
This paper assesses the merits of three different approaches to pixel-level human skin detection. The basis for the three approaches has been reported recently in the literature. The first two approaches [1, 2] use simple ratios and color space transforms respectively, whereas the third is a numerically efficient approach based on a 3-D RGB probability map, first implemented by Re-hg [3]. The Bayesian probabilities are made possible to compute only with the availability of a large appropriately labeled database. Over 12,000 images from the Compaq skin and non-skin databases [4] are used to quantitatively assess the three approaches. Thresholds are determined empirically to detect 95% of all skin-associate d pixels and assessment is then made in terms of the percentage of non-skin pixels incorrectly accepted. The lowest of these false acceptance rates is found to be about 20% given by the 3-D probability map.