Face Feature Detection for 3D Model of Talking Head with Speech Synthesis

Several new original algorithms for face feature detection are presented within this paper. The detected objects are used to produce realistic 3D model of human face. Presented methods have been tested and the results are discussed in the paper. The facial feature detection is based on human skin chromaticity and morphological characteristics of the human head. Output of the skin detection is used to isolate further important points of the analyzed face. The results of the exact feature detection lead to a more realistic 3D model of talking head used for multimedia speech synthesis.