Facial image indexing in multimedia databases

Pictures and video sequences showing human faces are of high importance in content-based retrieval systems, and consequently face detection has been e stablished as an important tool in the framework of many multimedia applica tions like indexing, scene classification and news summarisation, In this w ork, we combine skin colour and shape features with template matching in an efficient way for the purpose of facial image indexing. We propose an adap tive two-dimensional Gaussian model of the skin colour distribution whose p arameters are re-estimated based oil the current image or frame, reducing g eneralisation problems, Masked areas obtained from skin colour detection ar e processed using morphological tools and assessed using global shape featu res. The verification stage is based on a template matching variation provi ding robust detection. Facial images and video sequences are indexed accord ing to the number of included faces, their average colour components and th eir scale. leading to new types of content-based retrieval criteria in quer y-by-example frameworks. Experimental results have shown that the proposed implementation combines efficiency, robustness and speed, and could be easi ly embedded in generic visual information retrieval systems or video databa ses.