ADAPTIVE SKIN DETECTION USING MULTIPLE CUES (MP-P4)

Author(s) : Jinfeng Yang (Institute of Automation, Chinese Academy of Science, National Laboratory of Pattern Recognition, China) Zhouyu Fu (Institute of Automation, Chinese Academy of Science, National Laboratory of Pattern Recognition, China) Tieniu Tan (Institute of Automation, Chinese Academy of Science, National Laboratory of Pattern Recognition, China) Weiming Hu (Institute of Automation, Chinese Academy of Science, National Laboratory of Pattern Recognition, China)

Abstract : This paper presents an adaptive approach to skin detection using the multiple cues of several color spaces. First, we identify the skin cluster region using a closed curve based on our proposed nonlinear relationship among R, G and B components. Then we design a split machine which aids the extraction of the pixels with similar low-level features from an image. Finally, we build a nonlinear skin color classifier with an adaptive threshold chosen by analyzing the attributes of the extracted pixels under the HSL, YCbCr, YUV and YIQ color spaces. Experimental results show that our proposed method works very well in skin detection.

Menu